Zurac Sabina MD, PhD
University of Medicine and Pharmacy “Carol Davila” Bucharest
Histopathologic particularities of regression in melanoma
Authors: Sabina Zurac, Mirela Cioplea, Cristiana Popp, Luciana Nichita, Carmen Dumitru, Constantin Caruntu, Daniel Boda, Carolina Constantin, Monica Neagu
Melanoma is a malignant tumor with increasing incidence in young patients and high mortality rate; its prognosis varies depending on the stage of the tumor – thin melanomas (pT1) are usually controlled by surgical excision within oncologic limits, whereas metastatic tumors have general survival about 1 year. Following complex interactions between tumor cells and the host immune system, melanoma may present spontaneous tumor regression; recent studies are in favor of a favorable prognostic significance of regression; in this context, we studied regression as a natural model of tumor destruction.
The study included 98 patients with malignant melanoma, half with regression, half without regression. We analyzed the clinical data and consecrated histopathological prognostic factors in melanomas with regression (in areas with and without tumor regression) and in non-regression melanomas. We studied cell cycle / oncogene cycle control factors: p16, p21, cyclin D1 by IHC assays. The tumor cell positive for cell cycle control markers is focal in tumor cell nuclei, with no significant difference between melanomas with regression and those without regression.
We have studied the phenotype of inflammatory tumor infiltrate in melanoma with and without regression. The prototype of inflammatory infiltrate in cutaneous melanoma that has a favorable prognosis is composed of numerous CD3 + T lymphocytes, few or absent CD20 + B lymphocytes, few or absent plasma cells with CD138 + phenotype and present Langerhans cells with CD1a + or langerin + phenotype.
We studied the MMP expression of 1,2,3,9,11 and 13 and TIMP 1,2 and 3 in melanoma with and without regression. Our study showed a global decreased of MMP1 and MMP11 expression in the non-regressed component of melanomas with regression compared to melanomas without regression, thus possibly favoring less aggressive behavior in these cases. Also, regression in melanoma is correlated with diminished expression of MMP2, MMP3 and MMP11 in the non-regressed component of melanoma with regression than in the regressed component of the same tumors. Certain types of regression have associated specific alterations in the MMP expression. Melanoma with segmental regression showes a significantly lower expression of MMP1 in the nonregressed component compared to melanoma without regression. Melanoma with partial regression has MMP2 overexpression in the non-regressed component compared to the regressed component.
We also described the differences in the expression of TIMP1, TIMP2 and TIMP3 in the regressed and non-regressed components of regression melanoma. TIMP3 was overexpressed in all cases with segmental regression in the nonregressed component compared to the regressed component. Moreover, the trend towards overexpression of TIMP1 and TIMP2 in the non-regressed component in melanoma with partial regression as compared to melanomas without regression was evident. These results support the hypothesis that the morphological differences identified in the melanoma regression spectrum may correlate with prognosis, which could explain the prognostic differences reported in the literature for regression melanomas.
Acknowledgments: This work was partially supported by two grants of Romanian Ministery of Research and Innovation, UEFISCDI, PN II PT PCCA 2013 4 1407 (Project 190/2014) and PN-III-P1-1.2-PCCDI-2017-0341 (project 61PCCDI⁄2018).