Project 1: Pathology of the Sentinel and Non-Sentinel Lymph Nodes

点击翻译按钮获取中文摘要

Specific Aim 1. To permit completion of our detailed study of tumor burden and disposition in sentinel nodes (SN), we continue to recall slides from patients with positive SN. At the end of 2007, there were 148 sets of slides received for this purpose. The materials from 79 patients remain to be evaluated and we are actively seeking. Patient acquisition to trial MSLT-II is progressing favorably and we have reviewed slides from 917 primary melanomas (182 in 2007) and SN from 923 patients (208 in 2006). We still have some difficulty in obtaining immunohistology slides from some participating centers, but after direct contact with the participating pathologists this problem has been reduced. Contributing centers remain unenthusiastic about sending images digitally on the basis of cost and workload. Efforts continue to refine and simplify the arrangements for appropriate and sufficient evaluation of primary melanomas and sentinel lymph nodes and we have prepared detailed protocols for local and central handling of tissues. We continue our efforts to accrue additional contributing centers. As a number of these are off shore there is a continuing need to work closely with them to facilitate their presentation of trial documents to local committees that are not necessarily willing to review documents written in English. Attention to this problem continues to attract new centers in Finland, Germany and Sweden. Centers in the United Kingdom joined the Trial in 2007. An important area of activity was preparing the MSLT-II data for presentation the Data Safety Monitoring Board. Specific Aim 2: Lymphatic mapping and SN biopsy are well-established in the staging and management of patients with melanoma, and other malignancies that spread initially to the regional nodes. Identification of tumor in the SN is the most precise staging and prognostic technique currently available. The SN is the site of metastatic melanoma in between 16-20% of melanoma patients and if tumor is present in the SN it is customary to perform a complete dissection of the remaining regional nodes of the affected nodal basin, a combination of procedures that confers enhanced disease-free and melanoma-specific survival relative to patients treated by wide excision followed by a period of observation, with regional lymphadenectomy reserved for patients who develop clinically detectable regional nodal metastases (see Project III). This approach, while logical in the present state of our knowledge, almost certainly over treats some patients, as tumor is identified in the non-sentinel nodes (NSNs) of only one-third of SN-positive melanoma patients. If it were possible accurately to identify patients likely to have tumor in the NSNs, the remaining patients, with no tumor in the NSN, could be spared completion lymphadenectomy, a morbid operation that is unlikely to provide them clinical advantage. Characteristics of the primary tumor and the SN are promising sources of prediction of the presence of melanoma in the NSN. Micrometer-measured thickness, width of ulceration and mitotic rate are the most durable characteristic of the primary that correlates with presence of tumor in NSN, likelihood of subsequent recurrences and likelihood of death from melanoma (1). The presence and amount of tumor in lymph nodes [number of positive SN (1 v >1), p=0.0011, tumor area as a percentage of nodal area p=0.0016, diameter of the largest metastasis (<1 v >4mm), p=0.0012, tumor burden (%), and number of tumor foci, (<5 v >5), p=0.0042] accurately predict clinical outcome and the need for further surgery and/or adjuvant systemic therapy, however without sufficient precision to be totally accurate for the individual patient (2). Micrometer assessed relative tumor burden can serve as a surrogate for more cumbersome assessments of tumor area or volume (2) and there are some recent data that indicate that the location of tumor in the SN may be predictive, though in a review of data from MSLT-I patients, we have found this less significant than the above noted parameters (Appendix 1). Specifically, patients with SN tumor confined to the subcapsular sinus have been reported to be less likely to have tumor in the NSN develop tumor recurrence or die from melanoma. Attempts to repeat this attractive dispositional approach have nor been entirely successful (Scolyer, personal communication). Sentinel nodes show selective immune modulation, indexed as alterations of the relative area of nodal paracortex and area, frequency and dendritic morphology of paracortical dendritic leukocytes (3-5). Alterations are also noted in the activation phenotype of the high endothelial venules and in the transmigration of naive T lymphocytes through the walls of paracortical high endothelial venules. In 90 patients with a melanoma-positive SN who had a completion lymphadenectomy, we evaluated the capacity of these parameters to predict tumor in the NSN, the likelihood of melanoma recurrence and of death from melanoma (6). Sentinel nodes were bisected and sampled for histology and immunohistology using our standard protocol (7). We evaluated the micrometer-measured thickness of the primary (1) and the area of the node occupied by tumor (using a computer-assisted image analysis program: Planar Morphometry, Version 2.1, Southern Microlnstruments, Atlanta, GA) expressed as a percentage of the total area of the SN. In sections stained for S-100 protein, we assessed the area of the node occupied by paracortical interdigitating dendritic cells expressed as a percentage of total nodal area (4,7). We also determined the density of paracortical interdigitating dendritic cells/mm2 using the technique of Huang et al. (4). All 90 patients had a completion lymph node dissection and the NSN were evaluated for melanoma using HE-stained and immunostained sections (S-100 and HMB-45). We divided patients into those who remained tumor-free and those who recurred and died of melanomatosis. Median follow-up was 7 years (range 1-13 years). We recorded time from nodal surgery to recurrence/death and correlated primary thickness, tumor in the SN and the area and density of integrating dendritic cells in the SN with melanoma in the NSN and tumor recurrence and death from melanoma. Statistical techniques employed were the Wilcoxon Rank Sums, logistic regression; tree-based modeling, Cox proportional hazard regression modeling, the log rank test and Receiver Operating Characteristic curves. The relative area of tumor in the SN and Breslow thickness of the primary melanoma most accurately predicted tumor in the NSN (p=0.0001 in both cases - Wilcoxon Rank Sums). The presence of melanoma in the NSN was also predicted by the density of dendritic leukocytes in the paracortex (p=0.008 - Wilcoxon Rank Sums). These three observations assessed alone and in combination predict the presence of tumor in the NSN with high accuracy (Figures 1, 2). The same characteristics also correlated with tumor recurrence (tumor burden, p=0.0001, Breslow, p=0.0001 and dendritic cell density, p=0.0007) (Figure 3) and death from melanoma (tumor burden, p=0.0001, Breslow, p=0.0001 and dendritic cell density, p=0.0026) (Figure 2).