Aligning tumor mutational burden (TMB) quantification across diagnostic platforms: phase II of the Friends of Cancer Research TMB Harmonization Project.

BACKGROUND: Tumor mutational burden (TMB) measurements aid in identifying patients who are likely to benefit from immunotherapy; however, there is empirical variability across panel assays and factors contributing to this variability have not been comprehensively investigated. Identifying sources of variability can help facilitate comparability across different panel assays, which may aid in broader adoption of panel assays and development of clinical applications. MATERIALS AND METHODS: Twenty-nine tumor samples and 10 human-derived cell lines were processed and distributed to 16 laboratories; each used their own bioinformatics pipelines to calculate TMB and compare to whole exome results. Additionally, theoretical positive percent agreement (PPA) and negative percent agreement (NPA) of TMB were estimated. The impact of filtering pathogenic and germline variants on TMB estimates was assessed. Calibration curves specific to each panel assay were developed to facilitate translation of panel TMB values to whole exome sequencing (WES) TMB values. RESULTS: Panel sizes >667 Kb are necessary to maintain adequate PPA and NPA for calling TMB high versus TMB low across the range of cut-offs used in practice. Failure to filter out pathogenic variants when estimating panel TMB resulted in overestimating TMB relative to WES for all assays. Filtering out potential germline variants at >0% population minor allele frequency resulted in the strongest correlation to WES TMB. Application of a calibration approach derived from The Cancer Genome Atlas data, tailored to each panel assay, reduced the spread of panel TMB values around the WES TMB as reflected in lower root mean squared error (RMSE) for 26/29 (90%) of the clinical samples. CONCLUSIONS: Estimation of TMB varies across different panels, with panel size, gene content, and bioinformatics pipelines contributing to empirical variability. Statistical calibration can achieve more consistent results across panels and allows for comparison of TMB values across various panel assays. To promote reproducibility and comparability across assays, a software tool was developed and made publicly available.

Inactivation of misselected CD8 T cells by CD8 gene methylation and cell death.

Misselected CD8 cells that express T cell receptors (TCRs) that do not recognize class I major histocompatibility complex (MHC) protein can emerge from thymic selection. A postthymic quality control mechanism that purges these cells from the repertoire is defined here. The failure of mature CD8 cells to simultaneously engage their TCR and CD8 coreceptor triggers an activation process that begins with inhibition of CD8 gene expression through remethylation and concludes with up-regulation of surface Fas and Fas ligand and cellular apoptosis. Thus, inhibition of a death signal through continued TCR-CD8 coengagement of MHC molecules is a key checkpoint for the continued survival of correctly selected T cells. Molecular defects that prevent delivery of the death signal to mistakenly selected T cells underlie the expansion of double-negative T cells, which is the cellular signature of a subset of systemic autoimmune diseases.

Independent divergences in the CD4 binding site and V3 loop encoded in two seroprevalent Ugandan HIV-1 clinical isolates.

Two major epitopes expressed in HIV-1 have been recently shown to play a central role in virus neutralization. One of these important specificities is a type-specific or group-specific, principal neutralizing determinant (PND) located in the V3 loop of gp120. The other is a more broadly neutralizing determinant associated with the CD4 binding site. Structural and serological studies of the variation in these epitopes have become important in vaccine research. This report describes the analysis of the DNA clones encoding a region of gp120 that overlaps the V3 loop and the putative CD4 recognition site in two new African isolates, UG06c and UG23c. Phylogenetic analyses of the DNA sequences showed that the new African isolates clustered with two very distinct subtypes of HIV-1. UG06c was grouped with U455, D687, and Z321, previously classified as "HIV-1 subtype A" in the AIDS and human retroviruses database; and UG23c was grouped with MAL, JY1, NDK, ELI, and Z2Z6 classified as "HIV-1 subtype D." Considerable variation was apparent in the V3 loop. The divergence included the presence of the hexapeptides GP-GRSF and GLGQAL at the cap of the loop in UG06c and UG23c, respectively. The GPGR tetrapeptide in UG06c formed a beta-turn configuration similar to that of MN or IIIB. The beta-turn was not found to be a likely conformation for GLGQ. The amino acids previously implicated in CD4 binding and the associated neutralizing activity were relatively conserved. To assess a possible impact of the sequence and conformational variations on serological reactivity, UG06c and UG23c were subjected to neutralization assay.(ABSTRACT TRUNCATED AT 250 WORDS)

Phylogenetic and serological characterization of two Ugandan HIV-1 isolates.

HIV-1 isolates Ug06 and Ug23 were established in culture from peripheral blood mononuclear cells (PBMCs) of Ugandan subjects. The isolates were studied for phylogenetic and serological relationships with each other and with the laboratory strains, HTLV-IIIB and HIV-1MN. The results suggest that the Ugandan isolates are related to different subgroups of African viruses with 17.3% of genetic distance between UG06 and the U455 provirus (Uganda); and 12.6% of genetic distance between UG23 and the JY1 provirus (Zaire). Analysis of the predicted amino acid sequences for Ug06 and Ug23 showed marked sequence heterogeneity in the V3 region and CD4-binding site. A conserved amino acid sequence was identified in the C-terminal immunodominant region of the envelope glycoprotein gp120. The isolates were compared in virus-neutralization experiments with HTLV-IIIB and HIV-1MN stocks, using panels of Western blot-positive North American and Ugandan sera. The North American serum samples showed broad neutralizing activity against both of the Ugandan isolates. However, the Ugandan serum panel demonstrated strain-specific activity against either Ug06 or Ug23. Furthermore, the African serum specimens showed higher prevalence and titers of neutralizing activity against the HIV-1MN stock as compared with HTLV-IIIB.

Seroreactivity of analogous antigenic epitopes in glycoprotein 120 expressed in HIV-1 subtypes A, B, C, and D.

This article describes the impact of sequence variation on the distribution and seroreactivity of linear antigenic epitopes in gp120 encoded in new Ugandan HIV-1 clones from subtypes A, C, and D, and in North American clones from the B subtype. A region of the env gene encoding the C2 to V5 domains was PCR amplified from the lysates of peripheral blood leukocytes or from short-term cultured isolates. Computer-assisted analyses were conducted on the amino acid sequences to determine the distribution of surface structures in gp120. Despite marked sequence diversity, eight analogous epitopes were predicted for all clades of the virus analyzed. Synthetic peptides comprising the putative principal neutralizing determinant E2[V3], and other B cell epitopes E3[V3-V4], E4[V3-V4], E7[C3], and E8[V5], from a seroprevalent Ugandan isolate, AUG06c, were tested in ELISA for antigenicity with sera from Uganda, New York, and Thailand. Variable magnitudes of seroreactivity were observed for all of the peptides tested. However, a significantly higher degree of serum cross-reactivity was detected with the V3 loop peptide. ELISA reactivities of the same serum panel indicated that V3 loop peptides containing the apical residues GPGR (clones AUG06c and BRT3) or GPGQ (CUG045 and DUG044) were more antigenic and display extensive cross-reactivity as compared to analogous peptides comprising GLGQ (DUG23c), GQGQ (DUG042), or GPWG (BRT1). BETATURN analysis of the divergent V3 loop apical residues showed a good correlation of probable beta-turn occurrence with strong seroreactivity. These findings suggest that the major antigenic specificities in the divergent clades of HIV-1 are well conserved.(ABSTRACT TRUNCATED AT 250 WORDS)

Suppression of immune responses by CD8 cells. I. Superantigen-activated CD8 cells induce unidirectional Fas-mediated apoptosis of antigen-activated CD4 cells.

Stimulation of mature CD4 cells through the TCR induces cellular activation and expansion that are often followed by clonal elimination by a form of apoptosis termed activation-induced cell death. This process of CD4 cell apoptosis is generally thought to reflect clonal suicide and to be independent of other cell types. Here we show that during the response to the superantigen Staphylococcal enterotoxin A, activated CD8 cells, but not activated CD4 cells, suppress the CD4 proliferative response. Suppression by CD8 cells reflects their ability to induce CD4 cell apoptosis via ligation of Fas. Moreover, although activated CD8 cells that express Fas ligand and Fas eliminate CD4 cells through a Fas-dependent mechanism, they are themselves resistant to Fas-dependent apoptosis. These findings indicate a fundamental difference between the two major T cell subsets with regard to sensitivity to Fas-dependent apoptosis, expression of Fas ligand, and mediation of suppressive activity following immunization with superantigen.

Immunogenic potential of rgp120 from African HIV-1 subtype A.

Previous studies have shown that the African strains of HIV-1 mostly cluster with the subtypes A, C or D based on phylogenetic analysis of the ENV nucleotide sequences. In the present investigation we have examined the immunogenic potential of full length gp120 derived from the Ugandan HIV-1 subtype A isolate, AUG06c, using computer-based prediction methods and a plasmid-mediated immunization technique. Computer-assisted analysis of the amino acid residues identified 15 potential B-cell epitopes in gp120 of AUG06c. Despite marked variation in the primary sequences, these epitopes were shown to correspond well to analogous sites in gp120 derived from the subtype B reference clones, MN and IIIBBH10. The relative positions of the epitopes indicated that E9[V3], E14[C3] and E15[V5] correspond to the previously defined principal neutralizing determinant (PND) located in the V3 loop, the CD4 binding site and gp120 "immunodominant" region, respectively. Intramuscular inoculation of BALB/c mice with the ENV clones from AUG06c or from the subtype C clone, CUG045 elicited antibodies which react with the homologous but not with the heterologous PND peptide in ELISA. However, cocktail inoculation with the ENV plasmids from AUG06c and CUG045 elicited antibodies which reacted with both peptides. Antibody response to the other predicted epitopes of AUG06c was not as strong as the response to the PND peptide. The response of the mice to DNA-mediated immunization was further tested in a proliferation assay. Spleen cells derived from the immunized mice exhibited a strong proliferative response to homologous and heterologous PND peptides in [3H]thymidine incorporation assay. DNA-mediated immunization with rgp120 of AUG06c appears to elicit cellular immune response of relatively broad specificity.

Distribution of linear antigenic epitopes on GP120 encoded in sibling clones of novel New York HIV-1 subtype B isolates.

We have initiated studies to characterize the predominant subtypes of HIV-1 which account for infections in a defined cohort of intravenous (IV) drug addicts. A region of ENV encoding the C2 to the V5 regions was amplified from the leukocytes of two subjects currently enrolled in a methadone maintenance program at the Addiction Research and Treatment Corporation (ARTC), in Brooklyn, New York. This region of the viral genome encodes the principal neutralizing determinant (PND) located in the V3 loop, the immunogenic CD4-binding site, and six other linear antigenic epitopes in the envelope glycoprotein, gp120. Phylogenetic tree analysis of the nucleotide sequences showed that the sibling clones RT1.4, RT1.15, RT1.17, RT1.21 and RT3.6, RT3.10, RT3.11, RT3.12 and RT3.15 derived from the isolates, RT1 and RT3, respectively, cluster with "group B" viruses at 99% confidence level. Marked intra-patient and inter-patient sequence variation was apparent in the V3 loop. The divergence included the presence of a previously unreported hexapeptide GPWGTF at the cap of the loop in the clones from RT1. The North American consensus hexapeptide, GPGRAF, was identified in the cap of the loop from the clones of RT3. Four of the five sibling clones from RT3 were closely related whereas the other clone, RT3.15, displayed five amino acid mutations downstream of the V3 cap. To assess the effect of sequence variation on the distribution of linear antigenic epitopes, complementary computer software programs, were used to analyze the gp120 residues. Eight analogous antigenic epitopes were identified in the clones from both isolates despite the marked divergence in the primary sequences.(ABSTRACT TRUNCATED AT 250 WORDS)

A single point mutation in HIV-1 V3 loop alters the immunogenic properties of rgp120.

The results of the study presented in this report show that clones of env derived from genetically divergent HIV-1 field isolates fall into two major subsets based on the predicted secondary structure of the V3 region in gp120. One subset exemplified by the clones A-UG06c, B-RT3.12 and C-UG045 is predicted to assume a beta-turn conformation in the V3 loop and comprises the GPGX residues. The other subset exemplified by the clones D-UG23c and D-UG042 (GXGX) are deficient in the expression of the beta-turn in the loop. Since secondary conformations are highly likely to confer antigenic properties in a protein backbone at least for B cells, we have used nucleic acid immunization to test the effect of the beta-turn deficiency on the immunogenic potential of rgp120 encoded in these field isolates. As hypothesized, inoculation of BALB/c mice with the env plasmid encoding the beta-turn expressing rgp120 molecules resulted in the development of a vigorous antibody response to the homologous V3 loop peptides. In contrast, immunization with an rgp120 clone deficient in the beta-turn in the V3 loop showed no evidence of antibody development to the V3 loop. Instead, the latter clones triggered T cell proliferative responses and markedly increased the level of IL-2 and IFN-gamma production by T cells. Significantly, reconstitution of the beta-turn conformation by site-directed mutagenesis of a single V3 loop residue yielded rgp120 molecules which restored antibody production while diminishing the cell-mediated immune (CMI) responses to the V3 residue. These observations demonstrate the marked impact of a single amino acid substitution on the immunogenic properties of V3 region in gp120 encoded by divergent HIV-1 field isolates.

Relative reactivity of the V3 loop PND of HIV-1 subtypes A, B, C, D, and F with sera from selected Ugandan localities.

Synthetic peptides comprising the predicted principal neutralizing determinant (PND) in new African and North American HIV-1 clones were tested in ELISA for reactivity with ninety six serum samples from asymptomatic donors in six selected localities in Uganda. Irrespective of the geographical origin of the samples, the majority of the test sera cross-reacted at high intensities with the peptides derived from the North American clone, BRT3.6 (Group B), the Ugandan clone, CUG045, (Group C), and the Romanian clone, FRMA (Group F). The frequency of reactivity of the peptides from BRT3.6, CUG045, and FRMA were within the ranges of 57-100%, 50-100%, and 57-100%, respectively, for the sera collected from these disparate localities. In contrast to these findings, the V3 peptides derived from the other Ugandan isolates showed a more restricted pattern of reactivity with the same serum samples: AUG06c (1-63%), DUG23c (2%), and DUG044 (38-87%). The results from ELISA inhibition assay indicated that the V3 peptide from BRT 3.6, CUG045, and FRMA express closely related antigenic specificities quite distinct from those in AUG06c and DUG044. The residues comprising the PND in BRT 3.6, CUG045, and FRMA appear to be well conserved in the HIV-1 subtypes prevalent in the selected Ugandan locales.

PIP: In Uganda, health workers collected serum samples from 96 asymptomatic HIV-1 infected blood donors in Ishaka and Mbarara (southwest), Kisenyi and Kampala (central), and Lugazi and Jinja (east) so researchers working in a biochemistry laboratory at The City University of New York could describe the relative reactivity of the V3 loop from HIV-1 subtypes A, B, C, D, and F, as well as study the antigenic relationships within the PND encoded in the divergent HIV-1 subtypes. Regardless of geographic origin, the V3 peptides from most of the sera (at least 50%) collected in Uganda cross-reacted at high frequencies with the peptides derived from the novel North American clone (BRT3.6), the Ugandan clone (CUG045), and the Romanian clone (FRMA). The frequency of reactivity of these peptides with the test sera ranged from 57% to 100% for BRT3.6, from 50% to 100% for CUG045, and from 57% to 100% for FRMA. The V3 peptides from other Ugandan isolates (AUG06c, DUG044, and DUG23c) were less reactive with the same serum samples than BRT3.6, CUG045, and FRMA: 1-63%, 38-87%, and 2%, respectively. This finding suggests that the antigenic determinants expressed in AUG06c, DUG044, and DUG23c may not represent the PND encoded in most HIV-1 strains afflicting the Ugandan communities. The V3 peptides from BRT3.6, CUG045, and FRMA express closely related antigenic specificities altogether different from those in AUG06c and DUG044. The HIV-1 subtypes present in the selected Ugandan sites appear to effectively conserve the residues making up the PND in BRT3.6, CUG045, and FRMA.