T cell subtype profiling measures exhaustion and predicts anti-PD-1 response.

Anti-PD-1 therapy can provide long, durable benefit to a fraction of patients. The on-label PD-L1 test, however, does not accurately predict response. To build a better biomarker, we created a method called T Cell Subtype Profiling (TCSP) that characterizes the abundance of T cell subtypes (TCSs) in FFPE specimens using five RNA models. These TCS RNA models are created using functional methods, and robustly discriminate between naïve, activated, exhausted, effector memory, and central memory TCSs, without the reliance on non-specific, classical markers. TCSP is analytically valid and corroborates associations between TCSs and clinical outcomes. Multianalyte biomarkers based on TCS estimates predicted response to anti-PD-1 therapy in three different cancers and outperformed the indicated PD-L1 test, as well as Tumor Mutational Burden. Given the utility of TCSP, we investigated the abundance of TCSs in TCGA cancers and created a portal to enable researchers to discover other TCSP-based biomarkers.

Analytical Performance of an Immunoprofiling Assay Based on RNA Models.

As immuno-oncology drugs grow more popular in the treatment of cancer, better methods are needed to quantify the tumor immune cell component to determine which patients are most likely to benefit from treatment. Methods such as flow cytometry can accurately assess the composition of infiltrating immune cells; however, they show limited use in formalin-fixed, paraffin-embedded (FFPE) specimens. This article describes a novel hybrid-capture RNA sequencing assay, ImmunoPrism, that estimates the relative percentage abundance of eight immune cell types in FFPE solid tumors. Immune health expression models were generated using machine learning methods and used to uniquely identify each immune cell type using the most discriminatively expressed genes. The analytical performance of the assay was assessed using 101 libraries from 40 FFPE and 32 fresh-frozen samples. With defined samples, ImmunoPrism had a precision of ±2.72%, a total error of 2.75%, and a strong correlation (r2 = 0.81; P < 0.001) to flow cytometry. ImmunoPrism had similar performance in dissociated tumor cell samples (total error of 8.12%) and correlated strongly with immunohistochemistry (CD8: r2 = 0.83; P < 0.001) in FFPE samples. Other performance metrics were determined, including limit of detection, reportable range, and reproducibility. The approach used for analytical validation is shared here so that it may serve as a helpful framework for other laboratories when validating future complex RNA-based assays.

Genomic heterogeneity of ALK fusion breakpoints in non-small-cell lung cancer.

In lung adenocarcinoma, canonical EML4-ALK inversion results in a fusion protein with a constitutively active ALK kinase domain. Evidence of ALK rearrangement occurs in a minority (2-7%) of lung adenocarcinoma, and only ~60% of these patients will respond to targeted ALK inhibition by drugs such as crizotinib and ceritinib. Clinically, targeted anti-ALK therapy is often initiated based on evidence of an ALK genomic rearrangement detected by fluorescence in situ hybridization (FISH) of interphase cells in formalin-fixed, paraffin-embedded tissue sections. At the genomic level, however, ALK rearrangements are heterogeneous, with multiple potential breakpoints in EML4, and alternate fusion partners. Using next-generation sequencing of DNA and RNA together with ALK immunohistochemistry, we comprehensively characterized genomic breakpoints in 33 FISH-positive lung adenocarcinomas. Of these 33 cases, 29 (88%) had detectable DNA level ALK rearrangements involving EML4, KIF5B, or non-canonical partners including ASXL2, ATP6V1B1, PRKAR1A, and SPDYA. A subset of 12 cases had material available for RNA-Seq. Of these, eight of eight (100%) cases with DNA rearrangements showed ALK fusion transcripts from RNA-Seq; three of four cases (75%) without detectable DNA rearrangements were similarly negative by RNA-Seq, and one case was positive by RNA-Seq but negative by DNA next-generation sequencing. By immunohistochemistry, 17 of 19 (89%) tested cases were clearly positive for ALK protein expression; the remaining cases had no detectable DNA level rearrangement or had a non-canonical rearrangement not predicted to form a fusion protein. Survival analysis of patients treated with targeted ALK inhibitors demonstrates a significant difference in mean survival between patients with next-generation sequencing confirmed EML4-ALK rearrangements, and those without (20.6 months vs 5.4 months, P<0.01). Together, these data demonstrate abundant genomic heterogeneity among ALK-rearranged lung adenocarcinoma, which may account for differences in treatment response with targeted ALK inhibitors.

Hybrid capture and next-generation sequencing identify viral integration sites from formalin-fixed, paraffin-embedded tissue.

Although next-generation sequencing (NGS) has been the domain of large genome centers, it is quickly becoming more accessible to general pathology laboratories. In addition to finding single-base changes, NGS allows for the detection of larger structural variants, including insertions/deletions, translocations, and viral insertions. We describe the use of targeted NGS on DNA extracted from formalin-fixed, paraffin-embedded (FFPE) tissue, and show that the short read lengths of NGS are ideally suited to fragmented DNA obtained from FFPE tissue. Further, we describe a novel method for performing hybrid-capture target enrichment using PCR-generated capture probes. As a model, we captured the 5.3-kb Merkel cell polyomavirus (MCPyV) genome in FFPE cases of Merkel cell carcinoma using inexpensive, PCR-derived capture probes, and achieved up to 37,000-fold coverage of the MCPyV genome without prior virus-specific PCR amplification. This depth of coverage made it possible to reproducibly detect viral genome deletions and insertion sites anywhere within the human genome. Out of four cases sequenced, we identified the 5' insertion sites in four of four cases and the 3' sites in three of four cases. These findings demonstrate the potential for an inexpensive gene targeting and NGS method that can be easily adapted for use with FFPE tissue to identify large structural rearrangements, opening up the possibility for further discovery from archival tissue.

SLOPE: a quick and accurate method for locating non-SNP structural variation from targeted next-generation sequence data.

MOTIVATION: Targeted 'deep' sequencing of specific genes or regions is of great interest in clinical cancer diagnostics where some sequence variants, particularly translocations and indels, have known prognostic or diagnostic significance. In this setting, it is unnecessary to sequence an entire genome, and target capture methods can be applied to limit sequencing to important regions, thereby reducing costs and the time required to complete testing. Existing 'next-gen' sequencing analysis packages are optimized for efficiency in whole-genome studies and are unable to benefit from the particular structure of targeted sequence data. RESULTS: We developed SLOPE to detect structural variants from targeted short-DNA reads. We use both real and simulated data to demonstrate SLOPE's ability to rapidly detect insertion/deletion events of various sizes as well as translocations and viral integration sites with high sensitivity and low false discovery rate. AVAILABILITY: Binary code available at http://www-genepi.med.utah.edu/suppl/SLOPE/index.html

Panton-Valentine Leukocidin-positive methicillin-resistant Staphylococcus aureus lung infection in patients with cystic fibrosis.

BACKGROUND: Panton-Valentine Leukocidin-expressing (PVL+) methicillin-resistant Staphylococcus aureus (MRSA) is an emerging pathogen worldwide causing fatal necrotizing pneumonias in otherwise healthy individuals but has not been described in patients with cystic fibrosis (CF). Following two cases of patients with CF admitted with lung abscesses in association with PVL+ MRSA, we examined the incidence and the clinical characteristics of MRSA acquisition in our CF patient population. METHODS: Newly acquired MRSA isolates from patients with CF followed up at St. Louis Children's Hospital were analyzed for the presence of Panton-Valentine leukocidin coding region, clindamycin susceptibility, staphylococcal cassette chromosome (SCC) mec type, and multilocus sequence type. Medical records and pulmonary function studies at the time of MRSA isolation were reviewed. RESULTS: MRSA isolates from 40 CF patients were available for analysis. Six children (15%) had PVL+ MRSA infection. All PVL+ organisms were clindamycin susceptible. Patients who acquired a PVL+ organism were more likely to have a focal pulmonary infiltrate on chest radiograph, including cavitary lung lesions in two patients (p = 0.04), a markedly greater decline in FEV1 at the time of MRSA detection (p = 0.01), and a significantly higher WBC count (p = 0.04) and absolute neutrophil count (p = 0.04). These patients were more likely to be admitted for IV antibiotic therapy for respiratory illnesses (p < 0.01). CONCLUSIONS: We describe the emergence of PVL+ MRSA in our CF population in association with development of invasive lung infections including lung abscesses. Early identification and treatment of CF patients with newly acquired PVL+ MRSA may be crucial.

Expression of a bcr-1 isoform of RARalpha-PML does not affect the penetrance of acute promyelocytic leukemia or the acquisition of an interstitial deletion on mouse chromosome 2.

Expression of a bcr-3 isoform of retinoic acid receptor alpha-promyelocytic leukemia (RARalpha-PML) in mice expressing a bcr-1 isoform of PML-RARalpha is associated with increased penetrance of murine acute promyelocytic leukemia (APL) and the frequent acquisition of an interstitial deletion of one copy of mouse chromosome 2 (del(2)). To determine whether the isoform of RARalpha-PML is important for these effects, we created mice that expressed a bcr-1 isoform of RARalpha-PML. Coexpression with the bcr-1 isoform of PML-RARalpha did not increase the penetrance of APL (7 of 45 animals developed APL with PML-RARalpha alone vs 12 of 44 with both transgenes; P=.19). Furthermore, the frequency of del(2) in APL cells from doubly transgenic mice was not different from that of mice expressing PML-RARalpha alone (3 of 6 vs 6 of 12, respectively-P=1.38-compared with 11 of 11 for mice coexpressing PML-RARalpha and bcr-3 RARalpha-PML). The bcr-1 and bcr-3 isoforms of RARalpha-PML, therefore, have different biological activities that may be relevant for the pathogenesis of murine APL.

A Drosophila model of multiple endocrine neoplasia type 2.

Dominant mutations in the Ret receptor tyrosine kinase lead to the familial cancer syndrome multiple endocrine neoplasia type 2 (MEN2). Mammalian tissue culture studies suggest that RetMEN2 mutations significantly alter Ret-signaling properties, but the precise mechanisms by which RetMEN2 promotes tumorigenesis remain poorly understood. To determine the signal transduction pathways required for RetMEN2 activity, we analyzed analogous mutations in the Drosophila Ret ortholog dRet. Overexpressed dRetMEN2 isoforms targeted to the developing retina led to aberrant cell proliferation, inappropriate cell fate specification, and excessive Ras pathway activation. Genetic analysis indicated that dRetMEN2 acts through the Ras-ERK, Src, and Jun kinase pathways. A genetic screen for mutations that dominantly suppress or enhance dRetMEN2 phenotypes identified new genes that are required for the phenotypic outcomes of dRetMEN2 activity. Finally, we identified human orthologs for many of these genes and examined their status in human tumors. Two of these loci showed loss of heterozygosity (LOH) within both sporadic and MEN2-associated pheochromocytomas, suggesting that they may contribute to Ret-dependent oncogenesis.