Cellular cohesion as a prognostic factor in malignant melanoma: a retrospective study with up to 12 years follow-up

Abstract

Malignant melanomas have a high metastatic potential. Although the depth of tumour invasion is the single most important histological prognostic factor, in clinical practice this correlation is frequently challenged. In this study, we assessed the cohesion of malignant melanocytes in the dermal component of all primary melanomas in vertical growth phase with tumour thickness 40.76 mm diagnosed in our Department between 1990 and 1995. The rationale behind this morphological evaluation was based on the hypothesis that a change in the adhesion molecule profile of melanoma cells may manifest visually discrete changes in the way that cells group together in dermal aggregates. We used a dyscohesion score based on the proportion of invasive tumour occupied by dyscohesive neoplastic cells and assessed its clinical significance by correlating it with the incidence of recurrence, regional and distant metastases and survival of the patients. Follow-up was up to 12 years. We found that the degree of melanoma cell dyscohesion was associated with the probability of local recurrence or metastasis. This correlation was particularly significant when dyscohesion involving an area smaller than 25% (dyscohesion score 1) of the dermal component was compared to dyscohesion involving a larger area (scores 2-4). Melanomas in the latter group had significantly increased likelihood of recurrence or metastasis (Po0.025, log-rank test). This was particularly the case in T1-T3 melanomas (Po0.005). Similarly, T1-T3 melanomas with dyscohesion score 1 had a significantly higher survival rate (Po0.025). In the same cohort, both disease-free survival and survival were not significantly correlated to thickness, probably due to the limited number of cases. Finally, we showed that extent of dyscohesion was independent of Breslow thickness or tumour regression. We believe that estimation of melanoma cell dyscohesion is a reliable histological prognostic factor that may be appropriate in clinical practice.

DeepRePath: Identifying the Prognostic Features of Early-Stage Lung Adenocarcinoma Using Multi-Scale Pathology Images and Deep Convolutional Neural Networks

“…In our study, we found that DeepRePath extracted tumor necrosis and tumor cell patterns under a low-power view (original magnification, 100×). Tumor necrosis and discohesive tumor cells may be key features related to aggressive tumor behavior [35,36]. However, architectural features related to prognoses, such as micropapillary features, lymphovascular invasion, increased stroma, and tumor spread through air spaces, could not be identified in our model.…”

Effect of morphological preservation artifacts on the quality of immunohistochemical staining for ER and Her-2

“…Therefore the presence of retraction of tumour cell clusters from the stroma may have occurred, at least in some cases, during processing and not related to fixation. Cellular discohesion is a feature of poorly differentiated tumours (10). Distinction between cellular discohesion and artifactual cytoplasmic retraction can be made on the arrangement of the cell membranes of the adjacent cells.…”

“…Distinction between cellular discohesion and artifactual cytoplasmic retraction can be made on the arrangement of the cell membranes of the adjacent cells. The cell membranes of the neighboring discohesive cells are randomly arranged and show cytoplasmic processes (10). We did not include the tumors with cellular separation due to discohesion into the tumors with artifacts in order to separately identify the true artifactual retraction of tumour cells.…”