A rolling circle amplification screen for polyomaviruses other than BKPyV in renal transplant recipients confirms high prevalence of urinary JCPyV shedding.

OBJECTIVES: Multiple novel human polyomaviruses (HPyVs) have been discovered in the last few years. These or other, unknown, nephrotropic HPyVs may potentially be shed in urine. METHODS: To search for known and unknown HPyVs we investigated BKPyV-negative urine samples from 105 renal transplant recipients (RTR) by rolling circle amplification (RCA) analysis and quantitative JCPyV PCR. Clinical data was analysed to identify risk factors for urinary polyomavirus shedding. RESULTS: In 10% (11/105) of the urine samples RCA with subsequent sequencing revealed JCPyV, but no other HPyV sequences. Using quantitative JCPyV PCR, 24% (25/105) of the samples tested positive. Overall sensitivities of RCA of 44% (11/25) in detecting JCPyV in JCPyV DNA-positive urine and 67% (10/15) for samples with JCPyV loads >10,000 copies/ml can be assumed. Despite frequent detectable urinary shedding of JCPyV in our cohort, this could not be correlated with clinical risk factors. CONCLUSION: Routine urinary JCPyV monitoring in BKPyV-negative RTR without suspected polyomavirus-associated nephropathy might be of limited diagnostic value. As RCA works in a sequence-independent manner, detection of novel and known polyomaviruses shed in sufficient quantities is feasible. High-level shedding of HPyVs other than BKPyV or JCPyV in the urine of RTR is unlikely to occur.

Detection of cytomegalovirus (CMV) by real-time PCR in fecal samples for the non-invasive diagnosis of CMV intestinal disease.

BACKGROUND: Cytomegalovirus (CMV) infection of the gastrointestinal tract can cause CMV intestinal disease (CMV-ID), a severe complication in immunocompromised patients. Current gold standard for diagnosing CMV-ID requires the analysis of colon biopsies. Testing of fecal samples by CMV PCR might be a non-invasive diagnostic alternative, but data on this method is scarce. OBJECTIVES: To evaluate the use of quantitative CMV real-time PCR in fecal samples for diagnosing CMV-ID. STUDY DESIGN: Fecal samples and lower intestinal tract biopsies from 66 patients were analyzed by quantitative CMV PCR. To evaluate the diagnostic significance of CMV detection by PCR in fecal samples, patients were classified according to the etiology of their intestinal disease (based on results of endoscopy, histopathology and quantitative CMV DNA detection in biopsies) into three groups: "CMV-ID", "non-CMV-ID", and "equivocal". RESULTS: 10/66 fecal samples were tested positive by quantitative CMV PCR, but CMV DNA loads were low (range <1000-11,000copies/ml). CMV detection by PCR in fecal samples was positive in 8/12 patients of the CMV-ID group, resulting in a sensitivity of 67% for diagnosing CMV-ID. With two exceptions, fecal CMV PCR was negative in the non-CMV-ID group (45/47) indicating a good specificity (96%). Moreover, CMV DNA detection in feces was associated with high CMV DNA levels in intestinal biopsies. CONCLUSIONS: Negative CMV PCR results from fecal samples cannot exclude CMV-ID and thus have to be confirmed by analyzing intestinal biopsies. However, positive fecal PCR results are diagnostically useful and might help to circumvent invasive diagnostic procedures as endoscopy.

No human virus sequences detected by next-generation sequencing in benign verrucous skin tumors occurring in BRAF-inhibitor-treated patients.

Patients treated with BRAF inhibitors (e.g. vemurafenib), a novel targeted therapy for advanced melanoma harbouring certain BRAF mutations, develop numerous adverse cutaneous side effects, including skin tumors such as squamous cell carcinoma or non-malignant verruciform keratinocyte proliferations, termed 'BRAF-inhibitor-associated verrucous keratosis (BAVK) lesions'. These keratinocyte proliferations are believed to be caused by paradoxical hyperactivation of the MAPK pathway in cells with wild-type BRAF, but mutated RAS. However, due to the clinical and histological verruca-like appearance of these lesions, additional aetiologic cofactors, such as infectious agents (i.e. oncogenic viruses), might be suspected. Therefore, we performed 454 high-throughput sequencing of BAVK lesions from vemurafenib-treated patients on the transcript level to identify actively transcribed viral sequences of known [e.g. human papilloma viruses (HPV)] or even yet-unknown viruses. Next-generation sequencing did not identify transcripts of any human viruses out of 1 595 161 reads obtained from BAVK lesions of four patients. Nevertheless, all controls were recognized correctly, and the detection of sequences derived from the cutaneous microbiome (e.g. skin commensals and bacterial phages) confirmed the validity and sensitivity of the sequencing data. Our results are consistent with preliminary histological and immunohistochemical findings recently reported by others, who also failed to detect the expression of HPV proteins in BAVK. Although the patient number is limited and we cannot exclude the possibility of having missed a viral transcript of very low abundance, our study argues against a viral aetiology of BRAF-inhibitor-associated verruciform keratoses occurring under vemurafenib.

Next-generation sequencing fails to identify human virus sequences in cutaneous squamous cell carcinoma.

Nonmelanoma skin cancer (NMSC) shows a strongly increased incidence in solid organ transplant recipients (OTRs) and AIDS patients, suggesting an infectious etiology. The role of certain viruses, i.e., cutaneous human papillomaviruses (HPVs), in NMSC in immunosuppressed patients remains controversial. Merkel cell polyomavirus (MCPyV), which was recently identified using high-throughput sequencing, has been linked to cutaneous proliferations. Here, we aimed to identify novel or known viral sequences at the transcript level in cutaneous squamous cell carcinomas (SCCs) from OTR by using 454 high-throughput pyrosequencing, which can produce long reads (~400 bp) and thus is better suited for the analysis of unknown sequences than other sequencing platforms. cDNA libraries from three OTR SCC biopsies were generated and submitted to next-generation sequencing using a 454 platform. Bioinformatic analysis included digital transcriptome subtraction and--in parallel-reference mapping as an alternative way for depleting human sequences. All control sequences introduced for bioinformatics analysis were recovered correctly. Among 717,029 454-sequenced transcripts, nearly all identified viral reads were derived from phages. Bacterial sequences originated from the skin flora or environmental sources. Our study did not reveal any transcripts of known oncogenic or related unknown human viruses. These findings suggest that there is no abundant expression of known human viruses, or viruses with a high degree of homology to known human viruses, in cutaneous SCCs of OTR. Further studies are required to exclude the presence of viruses in NMSC, which cannot easily be identified on the basis of sequence homology to known viruses.

Comparison of the performance of direct fluorescent antibody staining, a point-of-care rapid antigen test and virus isolation with that of RT-PCR for the detection of novel 2009 influenza A (H1N1) virus in respiratory specimens.

Although infections with the novel pandemic 2009 influenza A (H1N1) virus (A/H1N1/2009) appeared to be relatively mild during the first summer of circulation ('off season'), there has been significant morbidity and hospitalization and several fatal cases. Thus, rapid detection of A/H1N1/2009 is crucial for efficient treatment and infection control measures. In contrast to seasonal influenza, where point-of-care (POC) rapid antigen tests and direct fluorescent antibody (DFA) staining ensure rapid detection, diagnosis of A/H1N1/2009 has so far been based on RT-PCR. This study retrospectively compared the performance of the Quidel QuickVue POC test, DFA staining and virus isolation with that of RT-PCR for A/H1N1/2009 detection in 526 respiratory specimens collected during the first wave of the outbreak from May to September 2009. A/H1N1/2009 was detected in 9.1% (48/526) of samples. One hundred and thirty-seven of the A/H1N1/2009 PCR-negative samples were additionally tested using a RealAccurate Respiratory RT-PCR panel, revealing other respiratory viruses (mainly entero/rhino- and adenoviruses) in 42.3% (58/137). All methods analysed detected A/H1N1/2009 with excellent specificity but different sensitivities (POC test: 18.2%; DFA staining: 38.7%; virus isolation: 45.7%). Therefore, the POC test was not suitable for diagnosis, detecting A/H1N1/2009 only if present in high concentrations (corresponding median Ct value=19.0; range=16.5-21.4). DFA staining was also able to detect A/H1N1/2009 in specimens with a lower virus concentration (median Ct value=24.0; range=16.5-29.8). Virus isolation, which was positive after a median time of 7.5 days, was too time-consuming. In summary, DFA staining is superior to POC testing and may be appropriate for patients expected to have a rather high level of virus replication. Nevertheless, in DFA-negative specimens, A/H1N1/2009 should be excluded by RT-PCR.