Neoantigen prediction and computational perspectives towards clinical benefit: recommendations from the ESMO Precision Medicine Working Group.

BACKGROUND: The use of next-generation sequencing technologies has enabled the rapid identification of non-synonymous somatic mutations in cancer cells. Neoantigens are mutated peptides derived from somatic mutations not present in normal tissues that may result in the presentation of tumour-specific peptides capable of eliciting antitumour T-cell responses. Personalised neoantigen-based cancer vaccines and adoptive T-cell therapies have been shown to prime host immunity against tumour cells and are under clinical trial development. However, the optimisation and standardisation of neoantigen identification, as well as its delivery as immunotherapy are needed to increase tumour-specific T-cell responses and, thus, the clinical efficacy of current cancer immunotherapies. METHODS: In this recommendation article, launched by the European Society for Medical Oncology (ESMO), we outline and discuss the available framework for neoantigen prediction and present a systematic review of the current scientific evidence. RESULTS: A number of computational pipelines for neoantigen prediction are available. Most of them provide peptide major histocompatibility complex (MHC) binding affinity predictions, but more recent approaches incorporate additional features like variant allele fraction, gene expression, and clonality of mutations. Neoantigens can be predicted in all cancer types with high and low tumour mutation burden, in part by exploiting tumour-specific aberrations derived from mutational frameshifts, splice variants, gene fusions, endogenous retroelements and other tumour-specific processes that could yield more potently immunogenic tumour neoantigens. Ongoing clinical trials will highlight those cancer types and combinations of immune therapies that would derive the most benefit from neoantigen-based immunotherapies. CONCLUSIONS: Improved identification, selection and prioritisation of tumour-specific neoantigens are needed to increase the scope of benefit from cancer vaccines and adoptive T-cell therapies. Novel pipelines are being developed to resolve the challenges posed by high-throughput sequencing and to predict immunogenic neoantigens.

Development of tumor mutation burden as an immunotherapy biomarker: utility for the oncology clinic.

Background: Treatment with immune checkpoint blockade (ICB) with agents such as anti-programmed cell death protein 1 (PD-1), anti-programmed death-ligand 1 (PD-L1), and/or anti-cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) can result in impressive response rates and durable disease remission but only in a subset of patients with cancer. Expression of PD-L1 has demonstrated utility in selecting patients for response to ICB and has proven to be an important biomarker for patient selection. Tumor mutation burden (TMB) is emerging as a potential biomarker. However, refinement of interpretation and contextualization is required. Materials and methods: In this review, we outline the evolution of TMB as a biomarker in oncology, delineate how TMB can be applied in the clinic, discuss current limitations as a diagnostic test, and highlight mechanistic insights unveiled by the study of TMB. We review available data to date studying TMB as a biomarker for response to ICB by tumor type, focusing on studies proposing a threshold for TMB as a predictive biomarker for ICB activity. Results: High TMB consistently selects for benefit with ICB therapy. In lung, bladder and head and neck cancers, the current predictive TMB thresholds proposed approximate 200 non-synonymous somatic mutations by whole exome sequencing (WES). PD-L1 expression influences response to ICB in high TMB tumors with single agent PD-(L)1 antibodies; however, response may not be dependent on PD-L1 expression in the setting of anti-CTLA4 or anti-PD-1/CTLA-4 combination therapy. Disease-specific TMB thresholds for effective prediction of response in various other malignancies are not well established. Conclusions: TMB, in concert with PD-L1 expression, has been demonstrated to be a useful biomarker for ICB selection across some cancer types; however, further prospective validation studies are required. TMB determination by selected targeted panels has been correlated with WES. Calibration and harmonization will be required for optimal utility and alignment across all platforms currently used internationally. Key challenges will need to be addressed before broader use in different tumor types.

RECK controls breast cancer metastasis by modulating a convergent, STAT3-dependent neoangiogenic switch.

Metastasis is the primary cause of cancer-related death in oncology patients. A comprehensive understanding of the molecular mechanisms that cancer cells usurp to promote metastatic dissemination is critical for the development and implementation of novel diagnostic and treatment strategies. Here we show that the membrane protein RECK (Reversion-inducing cysteine-rich protein with kazal motifs) controls breast cancer metastasis by modulating a novel, non-canonical and convergent signal transducer and activator of transcription factor 3 (STAT3)-dependent angiogenic program. Neoangiogenesis and STAT3 hyperactivation are known to be fundamentally important for metastasis, but the root molecular initiators of these phenotypes are poorly understood. Our study identifies loss of RECK as a critical and previously unknown trigger for these hallmarks of metastasis. Using multiple xenograft mouse models, we comprehensively show that RECK inhibits metastasis, concomitant with a suppression of neoangiogenesis at secondary sites, while leaving primary tumor growth unaffected. Further, with functional genomics and biochemical dissection we demonstrate that RECK controls this angiogenic rheostat through a novel complex with cell surface receptors to regulate STAT3 activation, cytokine signaling, and the induction of both vascular endothelial growth factor and urokinase plasminogen activator. In accordance with these findings, inhibition of STAT3 can rescue this phenotype both in vitro and in vivo. Taken together, our study uncovers, for the first time, that RECK is a novel regulator of multiple well-established and robust mediators of metastasis; thus, RECK is a keystone protein that may be exploited in a clinical setting to target metastatic disease from multiple angles.

Genetic determinants at the interface of cancer and neurodegenerative disease.

It has been hypothesized that oncogenesis and neurodegeneration may share common mechanistic foundations. Recent evidence now reveals a number of genes in which alteration leads to either carcinogenesis or neurodegeneration, depending on cellular context. Pathways that have emerged as having critical roles in both cancer and neurodegenerative disease include those involving genes such as PARK2, ATM, PTEN, PTPRD, and mTOR. A number of mechanisms have been implicated, and commonly affected cellular processes include cell cycle regulation, DNA repair, and response to oxidative stress. For example, we have recently shown that the E3 ubiquitin ligase PARK2 is mutated or deleted in many different human malignancies and helps drive loss on chromosome 6q25.2-27, a genomic region frequently deleted in cancers. Mutation in PARK2 is also the most common cause of juvenile Parkinson's disease. Mutations in PARK2 result in an upregulation of its substrate cyclin E, resulting in dysregulated entry into the cell cycle. In neurons, this process results in cell death, but in cycling cells, the result is a growth advantage. Thus, depending on whether the cell affected is a dividing cell or a post-mitotic neuron, responses to these alterations may differ, ultimately leading to varying disease phenotypes. Here, we review the substantial data implicating specific genes in both cancer and neurodegenerative disease.

Ferredoxin reductase affects p53-dependent, 5-fluorouracil-induced apoptosis in colorectal cancer cells.

Loss of p53 gene function, which occurs in most colon cancer cells, has been shown to abolish the apoptotic response to 5-fluorouracil (5-FU). To identify genes downstream of p53 that might mediate these effects, we assessed global patterns of gene expression following 5-FU treatment of isogenic cells differing only in their p53 status. The gene encoding mitochondrial ferredoxin reductase (protein, FR; gene, FDXR) was one of the few genes significantly induced by p53 after 5-FU treatment. The FR protein was localized to mitochondria and suppressed the growth of colon cancer cells when over-expressed. Targeted disruption of the FDXR gene in human colon cancer cells showed that it was essential for viability, and partial disruption of the gene resulted in decreased sensitivity to 5-FU-induced apoptosis. These data, coupled with the effects of pharmacologic inhibitors of reactive oxygen species, indicate that FR contributes to p53-mediated apoptosis through the generation of oxidative stress in mitochondria.

Cooperative effects of genes controlling the G(2)/M checkpoint.

It is believed that multiple effectors independently control the checkpoints permitting transitions between cell cycle phases. However, this has not been rigorously demonstrated in mammalian cells. The p53-induced genes p21 and 14-3-3sigma are each required for the G(2) arrest and allow a specific test of this fundamental tenet. We generated human cells deficient in both p21 and 14-3-3sigma and determined whether the double knockout was more sensitive to DNA damage than either single knockout. p21(-/-) 14-3-3sigma(-/-) cells were significantly more sensitive to DNA damage or to the exogenous expression of p53 than cells lacking only p21 or only 14-3-3sigma. Thus, p21 and 14-3-3sigma play distinct but complementary roles in the G(2)/M checkpoint, and help explain why genes at the nodal points of growth arrest pathways, like p53, are the targets of mutation in cancer cells.

High frequency of hypermethylation at the 14-3-3 sigma locus leads to gene silencing in breast cancer.

Expression of 14-3-3 final sigma (final sigma) is induced in response to DNA damage, and causes cells to arrest in G(2). By SAGE (serial analysis of gene expression) analysis, we identified final sigma as a gene whose expression is 7-fold lower in breast carcinoma cells than in normal breast epithelium. We verified this finding by Northern blot analysis. Remarkably, final sigma mRNA was undetectable in 45 of 48 primary breast carcinomas. Genetic alterations at final sigma such as loss of heterozygosity were rare (1/20 informative cases), and no mutations were detected (0/34). On the other hand, hypermethylation of CpG islands in the final sigma gene was detected in 91% (75/82) of breast tumors and was associated with lack of gene expression. Hypermethylation of final sigma is functionally important, because treatment of final sigma-non-expressing breast cancer cell lines with the drug 5-aza-2'-deoxycytidine resulted in demethylation of the gene and synthesis of final sigma mRNA. Breast cancer cells lacking final sigma expression showed increased number of chromosomal breaks and gaps when exposed to gamma-irradiation. Therefore, it is possible that loss of final sigma expression contributes to malignant transformation by impairing the G(2) cell cycle checkpoint function, thus allowing an accumulation of genetic defects. Hypermethylation and loss of final sigma expression are the most consistent molecular alterations in breast cancer identified so far.

PPARdelta is an APC-regulated target of nonsteroidal anti-inflammatory drugs.

PPARB was identified as a target of APC through the analysis of global gene expression profiles in human colorectal cancer (CRC) cells. PPARdelta expression was elevated in CRCs and repressed by APC in CRC cells. This repression was mediated by beta-catenin/Tcf-4-responsive elements in the PPARdelta promotor. The ability of PPARs to bind eicosanoids suggested that PPARdelta might be a target of chemopreventive non-steroidal anti-inflammatory drugs (NSAIDs). Reporters containing PPARdelta-responsive elements were repressed by the NSAID sulindac. Furthermore, sulindac was able to disrupt the ability of PPARdelta to bind its recognition sequences. These findings suggest that NSAIDs inhibit tumorigenesis through inhibition of PPARdelta, the gene for which is normally regulated by APC.

14-3-3Sigma is required to prevent mitotic catastrophe after DNA damage.

14-3-3Sigma is a member of a family of proteins that regulate cellular activity by binding and sequestering phosphorylated proteins. It has been suggested that 14-3-3sigma promotes pre-mitotic cell-cycle arrest following DNA damage, and that its expression can be controlled by the p53 tumour suppressor gene. Here we describe an improved approach to the generation of human somatic-cell knockouts, which we have used to generate human colorectal cancer cells in which both 14-3-3sigma alleles are inactivated. After DNA damage, these cells initially arrested in the G2 phase of the cell cycle, but, unlike cells containing 14-3-3sigma, the 14-3-3sigma-/- cells were unable to maintain cell-cycle arrest. The 14-3-3sigma-/- cells died ('mitotic catastrophe') as they entered mitosis. This process was associated with a failure of the 14-3-3sigma-deficient cells to sequester the proteins (cyclin B1 and cdc2) that initiate mitosis and prevent them from entering the nucleus. These results may indicate a mechanism for maintaining the G2 checkpoint and preventing mitotic death.

The neural cell adhesion molecules L1 and NCAM-180 act in different steps of neurite outgrowth.

The formation of neurocircuitry depends on the control of neurite outgrowth that, in turn, can be divided into two processes: nerve growth cone protrusion and neurite extension. It has long been known that the neural cell adhesion molecules L1 and NCAM-180 promote neurite outgrowth, but how they function in growth cones is unclear. We addressed the roles of L1 and NCAM-180 in neurite outgrowth by using microscale chromophore-assisted laser inactivation (micro-CALI) of these proteins to perturb their functions at precise times in single growth cones of embryonic chick dorsal root ganglion neurons grown in culture. Micro-CALI of L1 causes neurite retraction after a 10 min lag period but does not affect growth cone protrusion. In contrast, micro-CALI of NCAM-180 causes rapid growth cone retraction but does not affect neurite extension. The simultaneous inactivation of both these molecules resulted in both distinct effects that were segregated in time. The behavior of growth cones after these micro-CALI treatments resemble the drug-induced perturbation of microtubules for L1 and F-actin for NCAM-180. These findings suggest distinct roles in the growth cone for L1 and NCAM-180 in different steps of neurite outgrowth: L1 functions in neurite extension,whereas NCAM-180 functions in growth cone protrusion.

Mechanisms underlying nonsteroidal antiinflammatory drug-mediated apoptosis.

Nonsteroidal antiinflammatory drugs (NSAIDs) can inhibit colorectal tumorigenesis and are among the few agents known to be useful for the chemoprevention of neoplasia. Here, we show that the tumor suppressive effects of NSAIDs are not likely to be related to a reduction in prostaglandins but rather are due to the elevation of the prostaglandin precursor arachidonic acid (AA). NSAID treatment of colon tumor cells results in a dramatic increase in AA that in turn stimulates the conversion of sphingomyelin to ceramide, a known mediator of apoptosis. These results have significant implications for understanding and improving colon cancer chemoprevention.

Converting cancer genes into killer genes.

Over the past decade, it has become clear that tumorigenesis is driven by alterations in genes that control cell growth or cell death. Theoretically, the proteins encoded by these genes provide excellent targets for new therapeutic agents. Here, we describe a gene therapy approach to specifically kill tumor cells expressing such oncoproteins. In outline, the target oncoprotein binds to exogenously introduced gene products, resulting in transcriptional activation of a toxic gene. As an example, we show that this approach can be used to specifically kill cells overexpressing a mutant p53 gene in cell culture. The strategy may be generally applicable to neoplastic diseases in which the underlying patterns of genetic alterations or abnormal gene expression are known.

IVE (Image Visualization Environment): a software platform for all three-dimensional microscopy applications.

IVE (Image Visualization Environment) is a software platform designed from the outset to handle all aspects of modern computerized multidimensional microscopy. This platform provides users with an execution environment in which 5D data (XYZ, wavelength, and time) can be easily manipulated for the purpose of data collection, processing, display, and analysis. During the entire process, powerful data display functions are readily available for extracting complicated three-dimensional information through data visualization. By employing both the shared memory and multitasking features of the UNIX operation system, individual functions can be implemented as separate programs, and multiple programs can access the same data pool simultaneously. This enables users to combine the functionalities of different programs to facilitate each unique data analysis task. Furthermore, by defining an appropriate program execution model, commonly shared functional components such as data display, data I/O and user interface, etc. can be implemented using simple IVE library calls. This dramatically reduces the program development time and ensures consistency throughout the entire software system. As a result, users can quickly master the microscopy software system and new functions can be easily integrated, as different functional requirements arise for different research projects.

Identification of a gene encoding a novel protein-tyrosine kinase containing SH2 domains and ankyrin-like repeats.

Using the polymerase chain reaction with primers corresponding to conserved regions in the kinase domain of protein-tyrosine kinases, we amplified segments of several protein-tyrosine kinase genes from Hydra vulgaris, a member of the ancient metazoan phylum Cnidaria. Characterization of cDNA clones for one of these genes, HTK16, revealed that it encodes a non-receptor protein-tyrosine kinase with two SH2 domains but no SH3 domain. In this regard the predicted HTK16 protein resembles two mammalian non-receptor protein-tyrosine kinases, the products of the ZAP-70 and syk genes. However, the HTK16 protein contains five ankyrin-like repeats, a structural motif which has not previously been found in protein-tyrosine kinases. The HTK16 protein also contains a potential tyrosine phosphorylation site in its carboxyl-terminal tail which resembles the phosphorylation site in members of the src family. RNA hybridization analysis indicates that the HTK16 gene is expressed in epithelial cells, cells which also express the Hydra homologue of the src protein. Our finding of the HTK16 gene in Hydra indicates that diversification of genes encoding non-receptor protein-tyrosine kinases was a very early event in metazoan evolution.