Abstract

Gene expression profiles as well as genomic imbalances are correlated with disease progression in uveal melanoma (UM). We integrated expression and genomic profiles to obtain insight into the oncogenic mechanisms in development and progression of UM. We used tumor tissue from 64 enucleated eyes of UM patients for profiling. Mutations and genomic imbalances were quantified with digital PCR to study tumor heterogeneity and molecular pathogenesis. Gene expression analysis divided the UM panel into three classes. Class I presented tumors with a good prognosis and a distinct genomic make up that is characterized by 6p gain. The UM with a bad prognosis were subdivided into class IIa and class IIb. These classes presented similar survival risks but could be distinguished by tumor heterogeneity. Class IIa presented homogeneous tumors while class IIb tumors, on average, contained 30% of non-mutant cells. Tumor heterogeneity coincided with expression of a set of immune genes revealing an extensive immune infiltrate in class IIb tumors. Molecularly, class IIa and IIb presented the same genomic configuration and could only be distinguished by 8q copy number. Moreover, UM establish in the void of the immune privileged eye indicating that in IIb tumors the infiltrate is attracted by the UM. Combined our data show that chromosome 8q contains the locus that causes the immune phentotype of UM. UM thereby provides an unique opportunity to study immune attraction by tumors.

Highlights

  • Uveal melanoma (UM) is a rare ocular neoplasm characterized by GNAQ and GNA11 mutations [1, 2]

  • Unsupervised cluster analysis of gene expression data of ~16,000 unique genes in 64 primary UM subdivided the tumors into two distinct classes [22]

  • Class I tumors presented a good prognosis while class II tumors were correlated with a bad prognosis similar to what has been shown previously (Supplementary Figure S1) [6]

Read more

Summary

Introduction

Uveal melanoma (UM) is a rare ocular neoplasm characterized by GNAQ and GNA11 mutations [1, 2]. Monosomy of chromosome 3 has been the most described biomarker that predicts survival in UM patients [5]. Markers on chromosome 3 that underlie this unique subdivision are largely unknown but the chromosome 3 status divides UM genetically into two groups which have a good and a bad prognosis [6]. This finding is supported by clinical and histopathological markers [7]: UM containing epithelioid cells are associated with monosomy 3 and a bad prognosis while UM purely made up of spindle cells are rarely lethal [8, 9]. Www.impactjournals.com/oncotarget these tumor characteristics can be applied to precisely predict disease outcome, we do not yet understand the sequence of genetic events in the development of UM, nor the relationship between the genetic changes and tumor behavior

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call