Unlocking the therapeutic potential of primary tumor-draining lymph nodes
- 332 Downloads
Lymph nodes draining the primary tumor are essential for the initiation of an effective anti-tumor T-cell immune response. However, cancer-derived immune suppressive factors render the tumor-draining lymph nodes (TDLN) immune compromised, enabling tumors to invade and metastasize. Unraveling the different mechanisms underlying this immune escape will inform therapeutic intervention strategies to halt tumor spread in early clinical stages. Here, we review our findings from translational studies in melanoma, breast, and cervical cancer and discuss clinical opportunities for local immune modulation of TDLN in each of these indications.
KeywordsTumor-draining lymph node Local immunotherapy Cervical cancer Melanoma Breast cancer TIMO 2018
Conventional dendritic cell(s)
Immunogenic cell death
Tumor-negative lymph node
Tumor-positive lymph node
Lymph node resident dendritic cell(s)
Pathologic complete response
Plasmacytoid dendritic cell(s)
Sentinel lymph node(s)
Sentinel node biopsy
Tumor-draining lymph node(s)
Many complex processes are involved in the metastatic spread of cancer cells from the primary tumor to lymph nodes and distant organs. The sentinel lymph node (SLN) is the first node to receive lymphatic drainage from the primary lesion and is of great importance in initiating an effective anti-tumor immune response; it also constitutes a first line of defense against metastatic spread . For many malignancies, the presence of tumor cells in tumor-draining lymph nodes (TDLN), and in the SLN in particular, is a key prognostic factor and, in some cases, predicates the course of treatment . In some tumors, e.g., cervical cancer (CxCa) or oral cancer, a complete lymphadenectomy provides overall survival benefit [3, 4, 5, 6]. However, for other indications, such as melanoma  and breast cancer (BrC), this is not the case .
Tumor-draining lymph nodes as a target for immunotherapy
The main focus of current immunotherapeutic strategies is on targeting the microenvironment of primary tumors and/or metastatic lesions, most notably by checkpoint inhibitors. As therapeutic targets, TDLN, and SLN in particular, are relatively undervalued, and clinically under-utilized. They are, nonetheless, essential players in anti-tumor immunity. In this focused review, we will discuss the importance of, and clinical opportunities for, therapeutic targeting of TDLN, based on findings from pre-clinical and clinical studies carried out by our group.
In the TDLN, tumor-specific T-cell responses are initiated. Here, effective priming of cytotoxic CD8+ T cells takes place upon tumor-specific (neo)antigen recognition, presented by APC, including DC and macrophages . Although DC represent only a small population of all the immune cell subsets in the LN, they are crucial in initiating an effective immune response. In cancer, however, TDLN are under the influence of tumor-derived factors, such as extracellular vesicles , IL-6 , TGF-β , prostaglandin-E2 (PGE2) , and VEGF [12, 13]. As a result, DC are suppressed and acquire an immature and M2 macrophage-like phenotype, and will, therefore, not properly cross-present in TDLN . During tumor progression and prior to metastasis, TDLN undergo many additional profound alterations leading to invasion by cells derived from the primary tumor [1, 2, 15]. Such alterations include increased lymphangiogenesis, blood vessel remodeling, and increased chemokine and cytokine secretion, which can ultimately lead to changes in immune cell composition, resulting in a ‘tumor-supportive’ microenvironment, i.e., the pre-metastatic niche . Moreover, with the ability of tumor cells to evade immune surveillance by the upregulation of immunosuppressive ligands and downregulation of MHC class I-molecules, this can eventually lead to the metastatic growth of tumor cells that have reached the TDLN .
Thus, immune modulation of TDLN could generate effective tumor-specific T-cell responses and in this way prevent metastatic spread. Considering that only a minor fraction of systemically administered drugs reaches the TDLN , locally applied therapies may be more effective in counteracting immune suppression in TDLN. Based on immune profiling and ex vivo proof-of-concept studies, we have conducted and are currently conducting a number of clinical trials aimed at immune potentiation of the TDLN through local delivery of immune modulatory drugs.
Immune profiling of lymph nodes in cancer
Over the past 2 decades, our group has pioneered the flow cytometry-based immune profiling of TDLN in humans. In these studies, we employ a scraping method (i.e., we scrape the cutting surface of a bisected TDLN) to obtain viable leukocytes from the TDLN, which was shown not to interfere with diagnostic procedures . Compared to dissociation of the entire node, we found similar viabilities and phenotypic characteristics of T-cell and DC subsets in scrapes . In addition, using multiparameter (fluorescent) IHC, we are currently working on improving our understanding of the TDLN architecture and cellular networks by studying (co-)localization of diverse immune cell subsets in their microenvironment [19, 20, 21].
The influence of primary and invasive melanoma on conventional DC in SLN
Conventional dendritic cell subsets found in skin-draining lymph nodes
Most affected by 
Dermal dendritic cells
Circulation (LN resident)
Circulation (LN resident)
In melanoma, we observed a significant negative correlation between the activation state (based on CD83 expression) of DDC and LC in the SLN and primary tumor burden (Breslow thickness) . Interestingly, primary tumor burden was not shown to have a significant effect on either the frequency or activation state of LNDC subsets. However, the presence of SLN tumor metastases did have a significant impact on both the frequency and activation state of conventional LNDC, the latter showing a reverse correlation with the size of the metastasis (Table 1). This suggests that the primary melanoma can create a pre-metastatic niche in the TDLN by suppressing the activation states of migratory cDC subsets, which was shown to be associated with a shorter local recurrence-free survival. Subsequently, TDLN metastasis suppress LNDC which, interestingly, was shown to be associated with a worse distant recurrence-free survival . The latter indicates an essential role for conventional LNDC in the induction of effective systemic anti-tumor immunity.
Immune modulation of the melanoma SLN
The 10-year melanoma-specific survival of stage I and II melanoma patients, defined as any primary tumor without regional or distant metastases, ranges from 98 to 75% depending on risk factors, such as Breslow tumor depth and tumor ulceration. After tumor spread to the regional LN, the 10-year melanoma-specific survival can drop to as low as 24% in patients with stage IIID melanoma . The unmet medical need for many of these patients stems from the fact that there is no widely used adjuvant treatment available to reduce the chances of disease recurrence, although systemic treatment (neo-adjuvant, i.e., preceding complete lymph node dissection) with immune checkpoint inhibitors in patients who are at very high risk of recurrence (high-risk stage III) and treatment with dual BRAF and MEK inhibitors in patients with BRAF V600E or V600K mutated stage III melanoma, has shown to improve recurrence-free survival [30, 31, 32, 33], and has recently been approved by the FDA. For all other early stage patients, there is a “wait and see” approach after surgical removal of the primary lesion and SLN.
Immune modulation of TDLN in breast cancer
Comparable DC-targeting therapeutic approaches may be implemented in patients with BrC, since both melanoma and BrC drain to LN in the skin catchment area with comparable migratory and LN-resident DC subset distribution profiles. In BrC, neoadjuvant chemotherapy (NAC) is one of the treatment options. A pathologic complete response (pCR) upon NAC is an independent predictor for favorable clinical outcome in all molecular subtypes . Interestingly, T-cell infiltration in BrC holds predictive value for response to chemotherapy . Since certain cytostatic drugs can induce immunogenic cell death (ICD), leading to the release of tumor-associated antigens , there is a clear rationale to combine NAC with DC-potentiating strategies to optimize tumor-specific T-cell priming in the TDLN. An early study from 1999 already showed a favorable effect on patient survival of combined GM-CSF with NAC in patients with locally advanced BrC . Patients were treated with doxorubicin, cyclophosphamide (both agents known to induce ICD) and GM-CSF at three-weekly intervals. After a maximum of six cycles, patients underwent surgery and postoperative radiotherapy. We observed higher frequencies of mature DCs in the TDLN of these patients, suggesting that GM-CSF is able to improve patient outcome via DC recruitment and maturation, and a subsequent anti-tumor response . Interestingly, we have observed a similar relationship between hampered activation of LNDC and tumor involvement of SLN in patients with BrC as we previously reported in melanoma (van Pul et al. manuscript submitted). Therefore, in analogy to our clinical findings in melanoma, CpG-based local immune potentiation in combination with NAC may improve response rates in patients with BrC. This certainly deserves further (pre-)clinical exploration.
The role of TDLN in cervical cancer
In contrast to melanoma and BrC, CxCa is caused by a persistent infection with high-risk strains of the human papillomavirus (HPV), mainly HPV16 and HPV18. HPV-specific T cells  as well as T cells that target non-viral tumor-associated (neo-)antigens  have been detected in CxCa TDLN. As HPV-derived antigens are highly immunogenic, it is assumed that an immunosuppressive environment facilitates immune escape and thereby causes lymphatic spread.
CxCa is a locally invading disease and initially metastasizes to pelvic TDLN. The presence of LN metastases in patients with CxCa is a crucial prognostic factor . Importantly, survival benefit was observed for CxCa patients who underwent complete lymphadenectomy upon low-volume disease detection in the SLN, or even upon the removal of solely tumor-negative LN [3, 5], indicating the presence of an unfavorable immune microenvironment in CxCa-draining pelvic LN. To understand the cellular basis for this phenomenon and to find new immunotherapeutic targets that would allow immune stimulatory conversion of the TDLN microenvironment, we performed several studies in which we found various immune escape mechanisms exploited by CxCa.
The influence of PD-L1+ M2-like macrophages on cervical cancer progression
Interestingly, flow cytometric characterization of diverse immune cell subsets in TDLN of CxCa patients, showed that in contrast to melanoma and BrC, Langerhans cells were hardly present in CxCa LN. Although higher levels of CD1a+ DCs were present in tumor-positive LN (LN+) as compared to tumor-negative LN (LN−) , these cells might have been derived from recruited and tumor-converted monocytes rather than conventional migratory CD1a+ DC. Remarkably, we did not find evidence of decreased LNDC activation. These results point to the requirement for a different immunotherapeutic approach aimed at TDLN conditioning in CxCa, than the one tested and proposed for melanoma and BrC, respectively.
In addition to higher levels of CD1a+ DCs, elevated levels of activated CD8+ T cells in LN+ suggested immune activation . However, this activation was apparently overruled by a highly immunosuppressed microenvironment in LN+ compared to LN−, with high expression levels of the checkpoint molecules PD-1 and CTLA-4 on T cells and the presence of MDSC. Moreover, high rates of Tregs were observed in LN+, which correlated with the rates of M2-like CD14+PD-L1+ APC. A cytokine release profile consistent with an immune suppressive microenvironment was observed as well, with high IL-10, IL-6, TNFα, and low IFNγ expression. In a comparative study of all dissected cervical TDLN from five patients with CxCa, we found that immune suppression (identified as low CD8+ T cell/FoxP3+ Treg ratios) preceded actual metastasis, creating metastatic niches in the tumor-draining lymphatic catchment area . We hypothesize that primary tumors are able to recruit (possibly via the secretion of CCL2)  and polarize CD14+ monocytes into suppressive PD-L1+ M2-like macrophages [(co)-expressing CD14 and/or CD163] . These M2-macrophage-like cells, induced by tumor-derived factors, are incapable of stimulating proper CD8+ T-cell responses, favor Treg expansion, and facilitate tumor progression by the production of pro-angiogenic and pro-tumor-invasive factors [14, 49].
In aggregate, our findings support the clinical exploration of immunotherapies in CxCa aimed at converting the prevailing immunosuppressive conditions in the primary tumor and TDLN into an immune-activated tumor-targeting environment.
Modulating TDLN in cervical cancer
Theoretical advantages of low-dose, local immune potentiation in early stage cancer
1. Low(er) tumor load
2. Low(er) levels of immune suppression
3. Limited tumor heterogeneity: clonal neoantigens 
5. Single administration provides long-lasting protection 
7. Pre-empts the need for expensive systemic therapies
8. Off-the-shelf generally applicable
9. Leveraging a (sub-optimally) primed T-cell repertoire in the TDLN 
In conclusion, we believe that TDLN are of major importance in initiating a robust anti-tumor response upon immune modulating therapies and should be targeted by local delivery of immune modulatory agents. Evidence for this was provided by i.t. delivery of CTLA-4 blocking antibodies in a mouse model, showing equivalent tumor control to systemic administration with reduced side effects . Interestingly, Chamoto and colleagues observed absent anti-tumor efficacy of PD-1 blockade in a mouse model with TDLN ablation, and so demonstrated TDLN to be indispensable, even for an immune modulatory agent assumed to be primarily active in the tumor microenvironment . Fransen et al. recently confirmed these results and, importantly, showed equal in vivo anti-tumor efficacy of low dose locally injected anti-PD-1 and of systemically administered high-dose anti-PD-1 .
The rational design of future clinical trials targeting TDLN should encompass combinatorial use of immunotherapeutic agents, such as oncolytic viruses and/or immune checkpoint blocking antibodies. Moreover, it will likely not be limited to the cancer types discussed in this focused review, but may also be applied to other solid tumors proven amenable to immunotherapy, such as, e.g., lung cancer and head-and-neck cancer.
Graphical illustrations were drawn using the images from Servier Medical Art by Servier with slight modifications (http://www.smart.servier.com).
TDG drafted the original outline of the review. JR, BDK, and AMH wrote the paper. TDG and ESJ added content and edited the paper. Final approval was given by all authors.
Data discussed in this review were derived from studies funded by the Dutch Cancer Society (KWF Kankerbestrijding VU 2013-6015 and VU 2015-7864).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflicts of interest.
Ethical approval and ethical standards
Cell line authentication
- 8.Giuliano AE, Ballman KV, McCall L, Beitsch PD, Brennan MB, Kelemen PR et al (2017) Effect of axillary dissection vs no axillary dissection on 10-year overall survival among women with invasive breast cancer and sentinel node metastasis: The ACOSOG Z0011 (Alliance) Randomized Clinical Trial. JAMA 318:918–926CrossRefGoogle Scholar
- 13.Karpanen T, Egeblad M, Karkkainen MJ, Kubo H, Yla-Herttuala S, Jaattela M et al (2001) Vascular endothelial growth factor C promotes tumor lymphangiogenesis and intralymphatic tumor growth. Cancer Res 61:1786–1790Google Scholar
- 14.van de Ven R, Lindenberg JJ, Oosterhoff D, de Gruijl TD (2013) Dendritic Cell plasticity in tumor-conditioned skin: CD14(+) cells at the cross-roads of immune activation and suppression. Front Immunol. 4:403Google Scholar
- 26.Sluijter BJR, van den Hout MFCM, Koster BD, van Leeuwen PAM, Schneiders FL, van de Ven R et al (2015) Arming the melanoma sentinel lymph node through local administration of CpG-B and GM-CSF: recruitment and activation of BDCA3/CD141(+) dendritic cells and enhanced cross-presentation. Cancer Immunol Res 3:495–505CrossRefGoogle Scholar
- 34.Vuylsteke RJCLM, Molenkamp BG, Gietema HA, Van Leeuwen PAM, Wijnands PGJTB, Vos W et al (2004) Local administration of granulocyte/macrophage colony-stimulating factor increases the number and activation state of dendritic cells in the sentinel lymph node of early-stage melanoma. Cancer Res 64:8456–8460CrossRefGoogle Scholar
- 37.Koster BD, van den Hout MFCM, Sluijter BJR, Molenkamp BG, Vuylsteke RJCLM, Baars A et al (2017) Local adjuvant treatment with low-dose CpG-B offers durable protection against disease recurrence in clinical stage I-II melanoma: data from two randomized phase II trials. Clin Cancer Res 23:5679–5686CrossRefGoogle Scholar
- 38.Bonnefoi H, Litière S, Piccart M, MacGrogan G, Fumoleau P, Brain E et al (2014) Pathological complete response after neoadjuvant chemotherapy is an independent predictive factor irrespective of simplified breast cancer intrinsic subtypes: a landmark and two-step approach analyses from the EORTC 10994/BIG 1-00 phase III trial. Ann Oncol 25:1128–1136CrossRefGoogle Scholar
- 41.Honkoop AH, Luykx-de Bakker SA, Hoekman K, Meyer S, Meyer OW, van Groeningen CJ et al (1999) Prolonged neoadjuvant chemotherapy with GM-CSF in locally advanced breast cancer. Oncologist 4:106–111Google Scholar
- 46.Heeren AM, Koster BD, Samuels S, Ferns DM, Chondronasiou D, Kenter GG et al (2015) High and interrelated rates of PD-L1+CD14+ antigen-presenting cells and regulatory T cells mark the microenvironment of metastatic lymph nodes from patients with cervical cancer. Cancer Immunol Res 3:48–58CrossRefGoogle Scholar
- 50.(2018) Pembrolizumab OK’d for cervical cancer. Cancer Discov 8:904Google Scholar
- 54.Andrews A (2015) Treating with checkpoint inhibitors—figure $1 million per patient. Am Health Drug Benefits 8:9Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.