Esophageal squamous cell carcinoma (ESCC) is highly heterogeneous. Our understanding of full molecular and immune landscape of ESCC remains limited, hindering the development of personalised therapeutic strategies. To address this, we perform genomic-transcriptomic characterizations and AI-aided histopathological image analysis of 120 Chinese ESCC patients. Here we show that ESCC can be categorized into differentiated, metabolic, immunogenic and stemness subtypes based on bulk and single-cell RNA-seq, each exhibiting specific molecular and histopathological features based on an amalgamated deep-learning model. The stemness subgroup with signature genes, such as WFDC2, SFRP1, LGR6 and VWA2, has the poorest prognosis and is associated with downregulated immune activities, a high frequency of EP300 mutation/activation, functional mutation enrichment in Wnt signalling and the highest level of intratumoural heterogeneity. The immune profiling by transcriptomics and immunohistochemistry reveals ESCC cells overexpress natural killer cell markers XCL1 and CD160 as immune evasion. Strikingly, XCL1 expression also affects the sensitivity of ESCC cells to common chemotherapy drugs. This study opens avenues for ESCC treatment and provides a valuable public resource to better understand ESCC.

Actinic keratoses (AKs) are lesions of epidermal keratinocyte dysplasia and are precursors for invasive cutaneous squamous cell carcinoma (cSCC). Identifying the specific genomic alterations driving the progression from normal skin to skin with AK to skin with invasive cSCC is challenging because of the massive UVR-induced mutational burden characteristic at all stages of this progression. In this study, we report the largest AK whole-exome sequencing study to date and perform a mutational signature and candidate driver gene analysis on these lesions. We demonstrate in 37 AKs from both immunosuppressed and immunocompetent patients that there are significant similarities between AKs and cSCC in terms of mutational burden, copy number alterations, mutational signatures, and patterns of driver gene mutations. We identify 44 significantly mutated AK driver genes and confirm that these genes are similarly altered in cSCC. We identify azathioprine mutational signature in all AKs from patients exposed to the drug, providing further evidence for its role in keratinocyte carcinogenesis. cSCCs differ from AKs in having higher levels of intrasample heterogeneity. Alterations in signaling pathways also differ, with immune-related signaling and TGFβ signaling significantly more mutated in cSCC. Integrating our findings with independent gene expression datasets confirms that dysregulated TGFβ signaling may represent an important event in AK‒cSCC progression.

Human papillomaviruses (HPVs) cause most cervical cancers and an increasing number of anogenital and oral carcinomas, with most cases caused by HPV16 or HPV18. HPV hijacks host signalling pathways to promote carcinogenesis. Understanding these interactions could permit identification of much-needed therapeutics for HPV-driven malignancies. The Hippo signalling pathway is important in HPV+ cancers, with the downstream effector YAP playing a pro-oncogenic role. In contrast, the significance of its paralogue TAZ remains largely uncharacterised in these cancers. We demonstrate that TAZ is dysregulated in a HPV-type dependent manner by a distinct mechanism to that of YAP and controls proliferation via alternative cellular targets. Analysis of cervical cancer cell lines and patient biopsies revealed that TAZ expression was only significantly increased in HPV18+ and HPV18-like cells and TAZ knockdown reduced proliferation, migration and invasion only in HPV18+ cells. RNA-sequencing of HPV18+ cervical cells revealed that YAP and TAZ have distinct targets, suggesting they promote carcinogenesis by different mechanisms. Thus, in HPV18+ cancers, YAP and TAZ play non-redundant roles. This analysis identified TOGARAM2 as a previously uncharacterised TAZ target and demonstrates its role as a key effector of TAZ-mediated proliferation, migration and invasion in HPV18+ cancers.

The multifunctional protein, splicing factor, proline- and glutamine-rich (SFPQ) has been implicated in numerous cancers often due to interaction with coding and non-coding RNAs, however, its role in melanoma remains unclear. We report that knockdown of SFPQ expression in melanoma cells decelerates several cancer-associated cell phenotypes, including cell growth, migration, epithelial to mesenchymal transition, apoptosis, and glycolysis. RIP-seq analysis revealed that the SFPQ-RNA interactome is reprogrammed in melanoma cells and specifically enriched with key melanoma-associated coding and long non-coding transcripts, including SOX10, AMIGO2 and LINC00511 and in most cases SFPQ is required for the efficient expression of these genes. Functional analysis of two SFPQ-enriched lncRNA, LINC00511 and LINC01234, demonstrated that these genes independently contribute to the melanoma phenotype and a more detailed analysis of LINC00511 indicated that this occurs in part via modulation of the miR-625-5p/PKM2 axis. Importantly, analysis of a large clinical cohort revealed that elevated expression of SFPQ in primary melanoma tumours may have utility as a prognostic biomarker. Together, these data suggest that SFPQ is an important driver of melanoma, likely due to SFPQ–RNA interactions promoting the expression of numerous oncogenic transcripts.