Predicting clinical outcomes in neuroblastoma with genomic data integration

Background

Neuroblastoma is a heterogeneous disease with diverse clinical outcomes. Current risk group models require improvement as patients within the same risk group can still show variable prognosis. Recently collected genome-wide datasets provide opportunities to infer neuroblastoma subtypes in a more unified way. Within this context, data integration is critical as different molecular characteristics can contain complementary signals. To this end, we utilized the genomic datasets available for the SEQC cohort patients to develop supervised and unsupervised models that can predict disease prognosis.

Results

Our supervised model trained on the SEQC cohort can accurately predict overall survival and event-free survival profiles of patients in two independent cohorts. We also performed extensive experiments to assess the prediction accuracy of high risk patients and patients without MYCN amplification. Our results from this part suggest that clinical endpoints can be predicted accurately across multiple cohorts. To explore the data in an unsupervised manner, we used an integrative clustering strategy named multi-view kernel k-means (MVKKM) that can effectively integrate multiple high-dimensional datasets with varying weights. We observed that integrating different gene expression datasets results in a better patient stratification compared to using these datasets individually. Also, our identified subgroups provide a better Cox regression model fit compared to the existing risk group definitions.

Conclusion

Altogether, our results indicate that integration of multiple genomic characterizations enables the discovery of subtypes that improve over existing definitions of risk groups. Effective prediction of survival times will have a direct impact on choosing the right therapies for patients.

Reviewers

This article was reviewed by Susmita Datta, Wenzhong Xiao and Ziv Shkedy.

     

The cytotoxic T-cell response to H-1 minor histocompatibility antigen differences

H-1-specific cytotoxic T cells were generated in in vitro secondary cultures. Effectors were assayed on H-2 compatible, peritoneal exudate cell targets in a ⁵¹Cr release assay. Target-cell lysis appeared to be specific for the H-1 type of the stimulator cells. Effector cells were T cells since they expressed Thy 1.2 alloantigen and required H-2 compatibility between donors of the stimulator cells, responder cells, and target cells for efficient lysis. Peritoneal exudate cells were found to be efficient specific competitors in the cytotoxicity assay. There appeared to be no strict correlation between in vitro cytotoxic T-cell activity and mean skin graft rejection times for a number of minor H and H-2D differences.     

Network-based cancer genomic data integration for pattern discovery

Background

Next-generation sequencing (NGS) has profoundly changed the approach to genetic/genomic research. Particularly, the clinical utility of NGS in detecting mutations associated with disease risk has contributed to the development of effective therapeutic strategies. Recently, comprehensive analysis of somatic genetic mutations by NGS has also been used as a new approach for controlling the quality of cell substrates for manufacturing biopharmaceuticals. However, the quality evaluation of cell substrates by NGS largely depends on the limit of detection (LOD) for rare somatic mutations. The purpose of this study was to develop a simple method for evaluating the ability of whole-exome sequencing (WES) by NGS to detect mutations with low allele frequency. To estimate the LOD of WES for low-frequency somatic mutations, we repeatedly and independently performed WES of a reference genomic DNA using the same NGS platform and assay design. LOD was defined as the allele frequency with a relative standard deviation (RSD) value of 30% and was estimated by a moving average curve of the relation between RSD and allele frequency.

Results

Allele frequencies of 20 mutations in the reference material that had been pre-validated by droplet digital PCR (ddPCR) were obtained from 5, 15, 30, or 40 G base pair (Gbp) sequencing data per run. There was a significant association between the allele frequencies measured by WES and those pre-validated by ddPCR, whose p-value decreased as the sequencing data size increased. By this method, the LOD of allele frequency in WES with the sequencing data of 15 Gbp or more was estimated to be between 5 and 10%.

Conclusions

For properly interpreting the WES data of somatic genetic mutations, it is necessary to have a cutoff threshold of low allele frequencies. The in-house LOD estimated by the simple method shown in this study provides a rationale for setting the cutoff.

     

Accelerated clinical discovery using self-reported patient data collected online and a patient-matching algorithm

Patients with serious diseases may experiment with drugs that have not received regulatory approval. Online patient communities structured around quantitative outcome data have the potential to provide an observational environment to monitor such drug usage and its consequences. Here we describe an analysis of data reported on the website PatientsLikeMe by patients with amyotrophic lateral sclerosis (ALS) who experimented with lithium carbonate treatment. To reduce potential bias owing to lack of randomization, we developed an algorithm to match 149 treated patients to multiple controls (447 total) based on the progression of their disease course. At 12 months after treatment, we found no effect of lithium on disease progression. Although observational studies using unblinded data are not a substitute for double-blind randomized control trials, this study reached the same conclusion as subsequent randomized trials, suggesting that data reported by patients over the internet may be useful for accelerating clinical discovery and evaluating the effectiveness of drugs already in use.     

High-resolution mapping of a minor histocompatibility antigen gene on mouse chromosome 2

Minor histocompatibility (H) loci are significant tissue transplantation barriers but are poorly understood at the genetic and molecular level. We describe the construction of a high-resolution genetic map that positions a class II MHC-restricted minor H antigen locus and orders 12 other genes and genetic markers within the we-un interval of mouse Chromosome (Chr) 2. An intersubspecific backcross between 10.UW/Sn-H-3 ^b and CAST/Ei, an inbred stock of Mus musculus castaneus, was used for this purpose. A total of 1168 backcross mice were generated, and 71 we-un recombinants were identified. Significant compression of the genetic map in males versus females and transmission distortion of CAST-derived we, un, and A ^w genes were observed. Monoclonal T cell lines specific for two minor H alloantigens, Hd-1^a and Hd2^a, encoded by gene(s) that map to the we-un interval were used to antigen type the backcross mice. The results suggest the Hd-1^a and Hd-2^a antigens are most likely encoded by a single gene, now referred to as H-3b. The determined gene order is we-0.09±0.09-Itp-0.62±0.23-D2Mit77-0.26±0.15[Evi-4, Pcna, Prn-p]-0.26±0.15-Scg-1-0.44±0.19-[Bmp2a, D2Mit70]-0.09±0.09-[D2Mit19, D2Mit46]-1.59±0.36-D2Mit28-0.97±0.28-D2Lerl-1.50±0.35-H-3b-0.26±0.15-un (% recombination±1 SE). Because the average resolution of the backcross is 0.09 cM, the backcross panel should facilitate the physical mapping and molecular identification of a number of genes in this chromosome region.     

SV40 T antigen acts as a minor histocompatibility antigen of SV40 T antigen tolerant transgenic mice

The ability of normal mice to mount an SV40 T antigen-specific cytolytic T lymphocytes response when immunized in vivo with splenocytes from the SV40 T antigen transgenic 427-line mice and restimulated in vitro with SV40-transformed fibroblasts, or when immunized with SV40 and restimulated with 427-line splenocytes, was analyzed. Both immunization schemes resulted in an SV40 T antigen-specific immune response, indicating the presence of SV40 T antigen-positive cells in the spleens of these transgenic mice. Normal mice engrafted with skin from 427 donors showed no rejection of the graft. Thus, SV40 T antigen in transgenic 427-line mice is expressed on an undefined cell type in the spleen and acts as a tissue-specific minor histocompatibility antigen.     

Visualization-based discovery and analysis of genomic aberrations in microarray data

Background  

Chromosomal copy number changes (aneuploidies) play a key role in cancer progression and molecular evolution. These copy number changes can be studied using microarray-based comparative genomic hybridization (array CGH) or gene expression microarrays. However, accurate identification of amplified or deleted regions requires a combination of visual and computational analysis of these microarray data.     

A human autoreactive T cell line specific for minor histocompatibility antigen(s) isolated from a bone marrow-grafted patient

A human autoreactive T cell line named Bur-1 has been obtained from a woman 4 mo after an allogeneic bone marrow transplantation (BMT) from one of her HLA-identical brothers. The phenotype of the cell line is 100% T11+ and over 90% T4+, and the karyotype confirms its donor (male) origin. These donor T cells proliferate specifically in the presence of donor's peripheral blood monocytes (PBM) but not recipient's cells, and they kill specifically donor's but not recipient's Epstein-Barr virus (EBV)-induced lymphoblastoid cell lines (LCL). PBM from another HLA-identical brother and from several unrelated donors also stimulate Bur-1 cells, and EBV-induced LCL from the same donors are killed in cytotoxicity assays. All of these donors share HLA-DR5 or HLA-DRw11 (the major split of HLA-DR5) with Bur-1 cells. However, some but not all of the PBM sharing HLA-DR5 with Bur-1 cells are recognized. Therefore, in contrast with the previously described autoreactive T cells, Bur-1 cells are not directed against self-MHC antigens but rather recognize autologous minor histocompatibility (mH) antigens in the context of autologous HLA class II molecules. Because both male and female cells can be recognized, the reacting minor antigen could not be the male-specific HY antigen. It is suggested that autoreactivity against mH antigens can be observed in bone marrow-grafted patients due to the education of bone marrow donor precursors in the recipient thymus not allowing tolerance to autologous (donor) mH antigens not shared by the recipient.     

T cells targeted against a single minor histocompatibility antigen can cure solid tumors

T cells responsive to minor histocompatibility (H) antigens are extremely effective in curing leukemia but it remains unknown whether they can eradicate solid tumors. We report that injection of CD8(+) T cells primed against the immunodominant H7(a) minor H antigen can cure established melanomas in mice. Tumor rejection was initiated by preferential extravasation at the tumor site of interferon (IFN)-gamma-producing H7(a)-specific T cells. Intratumoral release of IFN-gamma had two crucial effects: inhibition of tumor angiogenesis and upregulation of major histocompatibility complex (MHC) class I expression on tumor cells. Despite ubiquitous expression of H7(a), dissemination of a few H7(a)-specific T cells in extralymphoid organs caused neither graft-versus-host disease (GVHD) nor vitiligo because host nonhematopoietic cells were protected by their low expression of MHC class I. Our preclinical model yields unique insights into how minor H antigen-based immunotherapy could be used to treat human solid tumors.     

Expression and purification of the minor histocompatibility antigen, HA-1H generated in Escherichia coli

The minor histocompatibility antigen HA-1H is a potential immunotherapeutic molecule. It can be used as a target for graft versus leukaemia reactions to eliminate residual HA-1H expressing leukaemic cells in patients following haemopoietic stem cell transplantation, whereby HA-1H primed donor cells can be transferred into a patient via adoptive immunotherapy. However, thus far only synthetic peptides corresponding to a HLA-A *0201 restricted HA-1H epitope have been used to generate HA-1H specific T cells. We are the first laboratory to clone, express and purify a region of HA-1H using an Escherichia coli expression system. The recombinant HA-1H protein was purified under denaturing conditions and the affinity purification tag removed using thrombin to remove non-specific amino acids. The 92 amino acid recombinant protein was characterised by mass spectrometry. Our rationale is that by using a recombinant HA-1H protein rather than peptide, HA-1H specific T cells may be generated from presentation of multiple HA-1H epitopes complexed in different HLA molecules. A multi-epitope approach may have wider applicability and maybe more effective at leukaemia control. The recombinant HA-1H protein may also be used as a research tool to identify novel CD4(+) helper T cell and CD8(+) cytotoxic T cell epitopes.     

Kinetics of the anti-recipient cytotoxic cell response of mice with minor histocompatibility antigen graft-vs-host disease

The kinetics of the anti-recipient cytotoxic cell response of spleen cells from mice undergoing graft-vs-host disease (GVHD) induced to minor histocompatibility antigens were studied. Two population of cytotoxic cells were identified. Cytotoxic T lymphocytes (CTL) were present in recipient spleens 2 and 3 wk after transplantation but disappeared from the spleens before the onset of clinical disease. Cytotoxic T lymphocyte precursors (CTLp) were first detected in recipient spleens 2 wk after transplantation and were present during clinical disease. CTL may function as effectors in GVHD induced to minor histocompatibility antigens.     

Tetrameric HLA class I-minor histocompatibility antigen peptide complexes demonstrate minor histocompatibility antigen-specific cytotoxic T lymphocytes in patients with graft-versus-host disease.

Graft-versus-host disease (GvHD) is a chief complication of allogeneic bone marrow transplantation. In HLA-identical bone marrow transplantation, GvHD may be induced by disparities in minor histocompatibility antigens (mHags) between the donor and the recipient, with the antigen being present in the recipient and not in the donor. Cytotoxic T lymphocytes (CTLs) specific for mHags of the recipients can be isolated from the blood of recipients with severe GvHD (ref. 3). A retrospective study demonstrated an association between mismatch for mHags HA-1, -2, -4 and -5 and the occurrence of GvHD in adult recipients of bone marrow from HLA genotypically identical donors. Tetrameric HLA-peptide complexes have been used to visualize and quantitate antigen-specific CTLs in HIV-infected individuals and during Epstein-Barr virus and lymphocytic choriomeningitis virus infections. Here we show the direct ex vivo visualization of mHag-specific CTLs during GvHD using tetrameric HLA-class and I-mHag HA-1 and HY peptide complexes. In the peripheral blood of 17 HA-1 or HY mismatched marrow recipients, HA-1- and HY-specific CTLs were detected as early as 14 days after bone marrow transplantation. The tetrameric complexes demonstrated a significant increase in HA-1- and HY-specific CTLs during acute and chronic GvHD, which decreased after successful GvHD treatment. HLA class I-mHag peptide tetramers may serve as clinical tools for the diagnosis and monitoring of GvHD patients.     

Diversity of alpha and beta subunits of T-cell receptors specific for the H4 minor histocompatibility antigen

 The mouse H4 minor histocompatibility antigen (HA) includes a single H2K^b-bound peptide that stimulates rejection of skin allografts and generation of cytolytic T lymphocytes (CTL). We evaluated the diversity of the CTL response to this single minor HA peptide by sequencing alpha and beta chains of T-cell receptors (Tcr) from H4-specific CTLs as a first step toward understanding the diversity of Tcrs specific for single minor HA. We selected H4-specific CTL clones from short-term lines that were restricted by K^b (19 clones), K^bm5 (7 clones), and K^bm11 (10 clones). Whereas multiple V_α and V_β family members were identified in the panel of K^b-restricted CTLs, five VB genes and one VA subfamily were predominant in CTLs derived from multiple individuals. Similar distributions were observed with K^bm5- and K^bm11-restricted CTLs, suggesting that these two mutants did not alter Tcr gene usage observable at the VA and/or VB gene levels. Negatively charged residues were present in the CDR3 regions of 12/13 unique K^b-restricted beta chains. A comparable observation was made with K^bm5-restricted CTLs, and the distributions of these residues among CDR3 positions were similar in the two CTL panels. However, the K^bm11 mutation dramatically altered the distribution of these residues resulting in their presence in positions 10–12 of CDR3 regions in all CTLs. These results indicate that the Tcrs expressed by CTLs specific for this single minor HA peptide are oligoclonal and characterized by the presence of negatively charged residues in beta CDR3 regions, the distribution of which is profoundly altered by the K^bm11 mutation.     

The H4b minor histocompatibility antigen is caused by a combination of genetically determined and posttranslational modifications

Minor histocompatibility (H) Ag disparities result in graft-vs-host disease and chronic solid allograft rejection in MHC-identical donor-recipient combinations. Minor H Ags are self protein-derived peptides presented by MHC class I molecules. Most arise as a consequence of allelic variation in the bound peptide (p) that results in TCR recognizing the p/MHC as foreign. We used a combinational peptide screening approach to identify the immune dominant H2K(b)-restricted epitope defining the mouse H4(b) minor H Ag. H4(b) is a consequence of a P3 threonine to isoleucine change in the MHC-bound peptide derived from epithelial membrane protein-3. This allelic variation also leads to phosphorylation of the H4(b) but not the H4(a) epitope. Further, ex vivo CD8(+) T lymphocytes bind phosphorylated Ag tetramers with high efficiency. Although we document the above process in the minor H Ag system, posttranslational modifications made possible by subtle amino acid changes could also contribute to immunogenicity and immune dominance in tumor immunotherapeutic settings.     

XY gonadal dysgenesis: Genetic heterogeneity based upon clinical observations,H-Y antigen status and segregation analysis

Summary Clinical observations and segregation analysis indicate that XY gonadal dysgenesis is characterized by genetic heterogeneity. In addition to the type inherited in X-linked recessive fashion, segregation analysis of other families suggested another type by revealing that the proportion of affected sibs did not differ from that expected on the basis of a male-limited autosomal recessive inheritance. Further heterogeneity may be deduced on the basis of coexisting campomelic dwarfism or possibly also renal parenchymal abnormalities. These observations of genetic heterogeneity must be considered when interpreting studies in which individuals with XY gonadal dysgenesis may or may not show H-Y antigen.     

Identification of a novel HLA-B60-restricted T cell epitope of the minor histocompatibility antigen HA-1 locus

The polymorphic minor histocompatibility Ag HA-1 locus encodes two peptides, HA-1(H) and HA-1(R), with a single amino acid difference. Whereas the immunogenicity of the HA-1(R) allele has not yet been shown, the nonameric HA-1(H) peptide induces HLA-A2-restricted cytotoxic T cells in vivo and in vitro. It is not known whether the mHag HA-1(H) or HA-1(R) associates with other HLA class I molecules. Therefore, the polymorphic regions of both HA-1 alleles were analyzed to identify HLA class I binding peptides that are properly processed by proteasomal degradation. Peptide binding analyses were performed for all nonameric HA-1(H/R) peptides for binding to nine HLA class I molecules with >10% prevalence in the Caucasian population and for seven nonameric/decameric HA-1(H/R) peptides predicted to bind to HLA-A3, -B14, and -B60. Only the nonameric KECVL(H)/(R)DDL and decameric KECVL(H)/(R)DDLL peptides showed strong and stable binding to HLA-B60. In vitro digestion of 29-aa-long HA-1 peptides by purified 20S proteasomes revealed proper cleavage at the COOH termini of both HLA-B60 binding HA-1(H) and HA-1(R) peptides. In subsequent analyses, dendritic cells pulsed with the nonameric HA-1(R) peptide did not induce CTLs that recognize the natural HLA-B60/HA-1(R) ligand. In contrast, dendritic cells pulsed with the nonameric HA-1(H) peptide induced IFN-gamma-secreting T cells specific for the natural HLA-B60/HA-1(H) ligand in three HLA-B60(+) HA-1(RR) individuals, demonstrating the immunogenicity of the HLA-B60/HA-1(H) ligand. In conclusion, this study shows a novel HLA-B60-restricted T cell epitope of the minor histocompatibility Ag HA-1 locus.     

Cutting edge: the minor histocompatibility antigen H60 peptide interacts with both H-2Kb and NKG2D

Minor histocompatibility Ags elicit cell-mediated immune responses and graft rejection in individuals receiving MHC-matched tissues. H60 represents a dominant Ag that elicits a strong CTL response in C57BL/6 mice immunized against BALB.B. An 8-aa peptide in the H60 protein is presented by H-2K(b) and this is recognized by the TCR as an alloantigen. The intact H60 glycoprotein is a ligand for the costimulatory NKG2D receptor that is expressed by activated CD8(+) T cells. Thus, H60 may provide both an allogeneic peptide and its own costimulation. We show that mutation of an H-2K(b)-binding anchor residue in the H60 peptide completely abrogates binding of H60 glycoprotein to NKG2D and a synthetic H60 peptide partially blocks the binding of NKG2D to its ligand. Ligands of the human NKG2D receptor are remarkably polymorphic, suggesting that these may also serve as minor histocompatibility Ags.