Ronald J.C.L.M. Vuylsteke

Sentinel lymph node tumor load: an independent predictor of additional lymph node involvement and survival in melanoma

BackgroundEven though 60% to 80% of melanoma patients with a positive sentinel lymph node (SLN) have no positive additional lymph nodes (ALNs), all these patients are subjected to an ALN dissection (ALND) with its associated morbidity. The aim of this study was to predict the absence of ALN metastases in patients with a positive SLN by using features of the primary melanoma and SLN tumor load.MethodsOf 71 SLN-positive patients, 52 had metastasis limited to the SLN (group 1), and 19 had ≥1 positive ALN after ALND (group 2). The tumor load of the SLN was assessed by measuring the total surface area by computerized morphometry. Breslow thickness, ulceration and lymphatic invasion of the primary tumor, and total SLN metastatic area were tested as covariates predicting the absence of positive ALNs.ResultsThe mean SLN metastatic area was 1.18 mm2 (group 1) and 3.39 mm2 (group 2) (P = .003) and was the only significant and independent factor after multivariate analysis (P = .02). None of the patients with both a Breslow thickness <2.5 mm and an SLN metastatic area <.3 mm2 had a positive ALN.ConclusionsSLN metastatic area can be used to predict the absence of positive ALNs in melanoma patients. In this study, patients with a Breslow thickness <2.5 mm and an SLN tumor load <.3 mm2 seemed to have no positive ALN and had excellent survival. We hypothesize that this subgroup might not benefit from ALND. Prospective larger trials, using this model and randomizing between ALND and no ALND, should confirm this hypothesis.

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

The initial tumor-draining lymph node, the sentinel lymph node, not only constitutes the first expected site of micrometastasis but also the first point of contact between tumor-associated antigens and the adaptive immune system. A tumor-induced decrease in the frequency and activation state of sentinel lymph node dendritic cells will impair the generation of effective antitumor T-cell responses and increase the likelihood of metastatic spread. Here, we demonstrate that intradermal administration of granulocyte macrophage-colony stimulating factor around the excision site of stage I primary melanoma tumors increases the number and activation state of dendritic cells in the paracortical areas of the sentinel lymph node and enhances their binding to T cells. We conclude that local treatment of melanoma patients with granulocyte macrophage-colony stimulating factor, before surgery, conditions the sentinel lymph node microenvironment to enhance mature dendritic cell recruitment and hypothesize that this may be more conducive to the generation of T-cell–mediated antitumor immunity.

Tumor-Specific CD8+ T Cell Reactivity in the Sentinel Lymph Node of GM-CSF–Treated Stage I Melanoma Patients is Associated with High Myeloid Dendritic Cell Content

Purpose: Impaired immune functions in the sentinel lymph node (SLN) may facilitate early metastatic events during melanoma development. Local potentiation of tumor-specific T cell reactivity may be a valuable adjuvant treatment option. Experimental Design: We examined the effect of locally administered granulocyte/macrophage-colony stimulating factor (GM-CSF) on the frequency of tumor-specific CD8+ T cells in the SLN and blood of patients with stage I melanoma. Twelve patients were randomly assigned to preoperative local administration of either recombinant human GM-CSF or NaCl 0.9%. CD8+ T cells from SLN and peripheral blood were tested for reactivity in an IFNγ ELISPOT assay against the full-length MART-1 antigen and a number of HLA-A1, HLA-A2, and HLA-A3–restricted epitopes derived from a range of melanoma-associated antigens. Results: Melanoma-specific CD8+ T cell response rates in the SLN were one of six for the control group and four of six for the GM-CSF-administered group. Only one patient had detectable tumor-specific CD8+ T cells in the blood, but at lower frequencies than in the SLN. All patients with detectable tumor-specific CD8+ T cells had a percentage of CD1a+ SLN-dendritic cells (DC) above the median (i.e., 0.33%). This association between above median CD1a+ SLN-DC frequencies and tumor antigen–specific CD8+ T cell reactivity was significant in a two-sided Fishers exact test (P = 0.015). Conclusions: Locally primed antitumor T cell responses in the SLN are detectable as early as stage I of melanoma development and may be enhanced by GM-CSF-induced increases in SLN-DC frequencies.

Feasibility of flowcytometric quantitation of immune effector cell subsets in the sentinel lymph node of the breast after cryopreservation.

The sentinel lymph node (SLN) is an emerging focus for immunological research in breast cancer. Cryopreservation of SLN single-cell suspensions allows for simultaneous phenotypic multi-parameter analyses and minimizes operator dependent variability. This is of particular importance for immunomonitoring of large multicenter trials. However, little data are available regarding the influence of cryopreservation on phenotypic characteristics of lymph node dendritic cells and T cells. In this study we assessed the feasibility of cryopreservation of viable SLN cell samples for flowcytometric analysis, by comparing quantitative analyses of SLN cell samples after freeze-thawing with direct analysis of fresh SLN cell samples. SLN were collected from nine breast cancer patients. From each SLN cell sample, half was used for immediate analysis and half was analyzed after cryopreservation and thawing. Conventional dendritic cell (cDC) and T cell subsets were quantified and phenotypically characterized by flow cytometry. The observed frequencies of both CD1a(+) and CD1a(-)CD11c(+)CD14(-) cDC subsets showed significant correlation between the fresh and frozen-thawed samples. Similar high correlations were found for CD83 and CD86 expression markers on the more frequent (>0.2%) CD1a(+) and CD1a(-)CD11c(+)CD14(-) cDC subset, but not on the low-frequency (<0.2%) CD1a(+)CD11c(+)CD14(+) cDC subset. CD4/CD8 T cell ratios were comparable and were significantly correlated pre- and post-freezing. Regulatory CD4(+)CD25(hi) T cell frequencies and their FoxP3 expression levels were significantly higher after freezing-thawing than in the freshly analyzed samples. Nevertheless, a highly significant correlation was found for both parameters pre- and post-freezing. Cryopreservation and thawing seems a valid and practical alternative to direct analysis of fresh viable lymph node cells, without introducing cryo-dependent variance between SLN samples. However, enumeration of low-frequency cell populations and assessment of their marker expression levels are less reliable after cryopreservation and should be assessed and considered in the design of each clinical trial.

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

Purpose: Although risk of recurrence after surgical removal of clinical stage I–II melanoma is considerable, there is no adjuvant therapy with proven efficacy. Here, we provide clinical evidence that a local conditioning regimen, aimed at immunologic arming of the tumor-draining lymph nodes, may provide durable protection against disease recurrence (median follow-up, 88.8 months). Experimental Design: In two randomized phase II trials, patients, diagnosed with stage I–II melanoma after excision of the primary tumor, received local injections at the primary tumor excision site within 7 days preceding re-excision and sentinel lymph node (SLN) biopsy of either a saline placebo (n = 22) or low-dose CpG type B (CpG-B) with (n = 9) or without (n = 21) low-dose GM-CSF. Results: CpG-B treatment was shown to be safe, to boost locoregional and systemic immunity, to be associated with lower rates of tumor-involved SLN (10% vs. 36% in controls, P = 0.04), and, at a median follow-up of 88.8 months, to profoundly improve recurrence-free survival (P = 0.008), even for patients with histologically confirmed (i.e., pathologic) stage I–II disease (P = 0.02). Conclusions: Potentially offering durable protection, local low-dose CpG-B administration in early-stage melanoma provides an adjuvant treatment option for a large group of patients currently going untreated despite being at considerable risk for disease recurrence. Once validated in a larger randomized phase III trial, this nontoxic immunopotentiating regimen may prove clinically transformative. Clin Cancer Res; 23(19); 5679–86. ©2017 AACR.

Small bowel perforation caused by advanced melanoma.

The incidence of melanoma has been increasing over the years and it remains, despite the heterogeneous survival for different stages, a disease with high mortality. Dissemination occurs primarily by the lymphatic route, followed by the hematogenous route. Gastrointestinal metastases do occur, but they are mainly intraluminal mucosal melanomas. Peritoneal or primary mucosal melanomas are rare. Only a few cases have been described of patients presenting with acute abdominal pain due to a melanoma. In this report we present a young patient with no prior health problems. Due to silent progression of disease at first, and secondarily avoidance of medical consultation, she finally presented to our emergency department with signs of intestinal perforation. In the operating theater a massive metastasis in the intestines with perforation was seen, as well as many smaller intra-abdominal and cutaneous lesions. Approximately 35 cm of jejunum had to be resected. Furthermore, the primary melanoma on the left forearm was excised and turned out to be in almost complete regression. Although initial recovery after surgery was good, the patient died only one month after presentation due to the advanced nature of her disease, which points to the devastating effect of undiagnosed melanoma and gastrointestinal metastasis. Since the melanoma incidence is rising, similar cases may present in the near future. This emphasizes the importance of proper full physical examination in patients with atypical abdominal symptoms.

Abstract 4692: Local adjuvant treatment of clinical stage I-II melanoma with low dose CpG-B and/or GM-CSF: Long-term follow-up of three randomized controlled phase II trials

Introduction: Currently, there is no widely used adjuvant treatment available to improve survival after surgical excision of localized melanoma. Here, we present the clinical outcome of patients who participated in three randomized phase-II trials and received low-dose local immunotherapy, which was shown to be safe and to boost loco-regional and systemic anti-melanoma T cell immunity. Patients and Methods: In three single-center, single-blinded, randomized and placebo (saline) controlled phase-II clinical trials, patients with early stage melanoma were treated with 1) Granulocyte/Macrophage-Colony Stimulating Factor (GM-CSF), 2) unmethylated CpG type-B oligodeoxynucleotide CpG7909 (CpG-B), and 3) CpG-B, alone or combined with GM-CSF, through 1-4 intradermal injections at the site of the primary melanoma excision scar, within 7 days preceding re-excision and sentinel lymph node (SLN) biopsy. For clinical follow-up analysis, all treated patients were grouped together (treated group n=36) as were the patients who received saline (saline group n=28). Results: 10-year recurrence-free survival rate in the treated group was 94% (95% CI 78-98) versus 48% (95% CI 21-71) in the saline group (P=0.005), hazard ratio (HR) for recurrence was 0.15 (95% CI 0.06-0.60) for the treated group. This apparent antitumor efficacy was in line with the observation upon pathological examination of less tumor positive SLN in the treated group (P=0.05). The 10-year distant recurrence-free survival rate in treated patients was also higher (94%, 78-98 versus 59%, 28-81, HR 0.22, 0.07-0.93; P=0.04). Conclusion: Local low-dose immunotherapy in patients with early-stage melanoma may offer durable protection against distant recurrences. These findings warrant further clinical exploration of this local non-toxic immune potentiating regimen. Citation Format: Bas D. Koster, Mari F. van den Hout, Berbel J. Sluijter, Barbara G. Molenkamp, Ronald J. Vuylsteke, Arnold Baars, Paul A. van Leeuwen, Rik J. Scheper, Monique P. van den Tol, Alfons J. van den Eertwegh, Tanja D. de Gruijl. Local adjuvant treatment of clinical stage I-II melanoma with low dose CpG-B and/or GM-CSF: Long-term follow-up of three randomized controlled phase II trials [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4692. doi:10.1158/1538-7445.AM2017-4692

Abstract B138: Single low-dose administration of anti-CTLA-4/tremelimumab at the tumor excision site boosts systemic antitumor immunity in early-stage melanoma patients: Results from a Phase-1 study

While systemic immune checkpoint blockade is unsurpassed in the induction of clinically effective T cell immunity in advanced-stage melanoma patients, the risk of autoimmune-related side effects precludes its application in earlier stages. In a Phase-1 dose escalation study we have studied the safety, feasibility and immunological effects of intradermal delivery of a single low dose of anti-CTLA-4/tremelimumab at the primary tumor excision site of patients with clinically Stage I-II melanoma. Dose escalation (4x3 cohorts of 2, 5, 10 and 20 mg) has been concluded with one additional patient enrolled in a 20 mg expansion cohort. Breslow thickness ranged from 0.9 to 6 mm (mean 1.98 mm). Examination of the sentinel lymph node (SLN) revealed (micro-)metastases in five of the patients. The tremelimub administration was found to be safe without apparent side effects except for one case of localized vitiligo at the highest (20 mg) dose level. Of note, potent immunological effects were observed in equal measure at all dose levels. In all patients a decrease in systemic rates of activated Tregs (aTregs, defined as CD4+CD25hiFoxP3hi T cells) was observed after 7 days of treatment (p Specific T cell reactivity against long peptides derived from the melanoma antigens NY-ESO-1 and MART-1 was tested after in vitro re-stimulation in an IFNγ Elispot assay. In 7/13 patients evidence was found of tremelimumab-induced increases in systemic T cell rates, which remained elevated up to 12 weeks post-treatment. Remarkably, 5/13 patients showed pre-existent NY-ESO-1 reactivity, which in all cases had decreased by day 7, and in 3/5 patients had increased over baseline levels by day 21. T cell reactivity to either NY-ESO-1 or MART-1 was observed in 4/4 tested SLN. Of note, both NY-ESO-1 and MART-1 reactive T cells were detected in the SLN of the patient experiencing an isolated incidence of vitiligo after receiving 20 mg tremelimumab. In conclusion, local administration of a single low dose of anti-CTLA-4/tremelimumab has proven safe and effective in terms of attenuating aTreg levels in peripheral blood and boosting systemic antitumor immunity. As such it may offer an attractive, and up to now sorely lacking, adjuvant treatment option for early-stage melanoma patients at risk for tumor recurrence. Citation Format: Anita G.M. Stam, Kim M. van Pul, Bas D. Koster, Dafni Chondronasiou, Sinead M. Lougheed, M. Petrousjka van den Tol, Karin Jooss, Ronald J.C.L.M. Vuylsteke, Alfons J.M. van den Eertwegh, Tanja D. de Gruijl. Single low-dose administration of anti-CTLA-4/tremelimumab at the tumor excision site boosts systemic antitumor immunity in early-stage melanoma patients: Results from a Phase-1 study. [abstract]. In: Proceedings of the CRI-CIMT-EATI-AACR Inaugural International Cancer Immunotherapy Conference: Translating Science into Survival; September 16-19, 2015; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(1 Suppl):Abstract nr B138.

Adjuvant treatment of early-stage melanoma by local i.d. administration of low-dose CpG-B and GM-CSF increases recurrence-free survival: long-term follow-up of three randomized clinical trials

Meeting abstracts Currently, there is no adjuvant treatment available that has been shown to reduce the chances of recurrence after surgical excision of localized melanoma. Between 2002 and 2007 we conducted three randomized and placebo (saline) controlled Phase II clinical trials in which we

Counteracting breast-cancer induced immune suppression by reactivating lymph node-resident conventional dendritic cells (LNR-cDC)

Breast cancer (BrC)-derived factors inhibit proper development and activation of dendritic cells (DC), which in turn promote the expansion of regulatory T cells (Tregs), thus contributing to tumor progression and spread. In two separate BrC sentinel lymph node (SLN) cohorts, one from the EACRI in Portland and one from the VUmc in Amsterdam, we found metastasis-related immune suppression of CD11c+CD14-lymph node-resident conventional DC (LNR-cDC). Genome-wide microarray analyses on FACS sorted migratory CD1a+ cDC, LNR-cDC and CD14+ LNR-cDC isolated from breast-draining LN were performed. A focused analysis of c-type lectin receptors (CLR) and transcription factor genes associated with Ag uptake and cross-presentation, showed LNR-cDC to express relatively high transcript levels of Interferon-regulatory factor 8 (IRF8), chemokine receptor 1 (XCR1), BDCA3/thrombomodulin, and, remarkably, of Langerin, CLEC7A and CLEC4A. These are interesting observations, since these are all factors previously linked to efficient antigen cross-presentation and cross-priming of CD8+ T cells. The EACRI cohort (n=38) included LN from ductal carcinoma in situ (DCIS), tumor-negative SLN (SLN-) and metastasis-positive SLN (SLN+), analyzed with a 12-parameter flow cytometry panel. The VUmc cohort included healthy, prophylactically removed breast LN (HLN, n=16), SLN- and SLN+ (n=40). In both cohorts, there was significantly reduced activation of LNR-cDC, but not of migratory cDC, in SLN-versus DCIS/HLN with further reductions in SLN+. Notably, reduction in activation of LNR-cDC went hand-in-hand with increased Treg frequencies. We next explored the possibility to counteract this BrC-related suppression in SLN ex vivo. The combined activity of CpG-B and the JAK2/STAT3 inhibitor (STAT3i) AG490 resulted in increased activation of LNR-plasmacytoid DC and LNR-cDC. It also promoted production of IFNγ by CD8+ T cells, reduced production of IL-4 by CD4+ T cells, and decreased rates of suppressive Tregs, which were slightly elevated by CpG-B alone. Finally, the combination of CpG-B and STAT3i increased tumor-specific effector T cell responses as measured by IFNγ ELISPOT assay after in vitro restimulation against a pool of overlapping 15-mer peptides, covering the sequence of the BrC–associated antigen Mammaglobin-A. Combined, our data show that BrC-mediated suppression in SLN is primarily directed against the LNR-cDC subset (with particular cross-presenting characteristics and abilities), rather than CD1a+ cDC subsets that migrate to the SLN from local tissues. Stimulation with CpG-B and STAT3i could reactivate these suppressed LNR-cDC ex vivo, thereby restoring type-1 mediated anti-tumor immunity. In view of increasing evidence that immune-regulated pathways influence response to (neo)adjuvant chemotherapy, we anticipate clinical benefit of combination therapy with this immune-stimulatory cocktail.