Susanna K Tan1, Peidong Shen2, Martina I Lefterova3, Malaya K Sahoo3, Eula Fung2, Justin I Odegaard3, Ronald W Davis2, Benjamin A Pinsky1,3, Curt Scharfe4. 1. Department of Medicine, Division of Infectious Diseases and Geographic Medicine, Stanford University School of Medicine, Stanford, CA. 2. Stanford Genome Technology Center, Stanford University, Palo Alto, CA. 3. Department of Pathology, Stanford University School of Medicine, Palo Alto, CA. 4. Department of Genetics, Yale University School of Medicine, New Haven, CT.
Abstract
BACKGROUND: Viral infections are a major cause of complications and death in solid organ and hematopoietic cell transplantation. METHODS: We developed a multiplex viral sequencing assay (mVseq) to simultaneously detect 20 transplant-relevant DNA viruses from small clinical samples. The assay uses a single-tube multiplex PCR to amplify highly conserved virus genomic regions without the need for previous virus enrichment or host nucleic acid subtraction. Multiplex sample sequencing was performed using Illumina MiSeq, and reads were aligned to a database of target sequences. Analytical and clinical performance was evaluated using reference viruses spiked into human plasma, as well as patient plasma and nonplasma samples, including bronchoalveolar lavage fluid, cerebrospinal fluid, urine, and tissue from immunocompromised transplant recipients. RESULTS: For the virus spike-in samples, mVseq's analytical sensitivity and dynamic range were similar to quantitative PCR (qPCR). In clinical specimens, mVseq showed substantial agreement with single-target qPCR (92%; k statistic, 0.77; 259 of 282 viral tests); however, clinical sensitivity was reduced (81%), ranging from 62% to 100% for specific viruses. In 12 of the 47 patients tested, mVseq identified previously unknown BK virus, human herpesvirus-7, and Epstein-Barr virus infections that were confirmed by qPCR. CONCLUSIONS: Our results reveal factors that can influence clinical sensitivity, such as high levels of host DNA background and loss of detection in coinfections when 1 virus was at much higher concentration than the others. The mVseq assay is flexible and scalable to incorporate RNA viruses, emerging viruses of interest, and other pathogens important in transplant recipients.
BACKGROUND: Viral infections are a major cause of complications and death in solid organ and hematopoietic cell transplantation. METHODS: We developed a multiplex viral sequencing assay (mVseq) to simultaneously detect 20 transplant-relevant DNA viruses from small clinical samples. The assay uses a single-tube multiplex PCR to amplify highly conserved virus genomic regions without the need for previous virus enrichment or host nucleic acid subtraction. Multiplex sample sequencing was performed using Illumina MiSeq, and reads were aligned to a database of target sequences. Analytical and clinical performance was evaluated using reference viruses spiked into human plasma, as well as patient plasma and nonplasma samples, including bronchoalveolar lavage fluid, cerebrospinal fluid, urine, and tissue from immunocompromised transplant recipients. RESULTS: For the virus spike-in samples, mVseq's analytical sensitivity and dynamic range were similar to quantitative PCR (qPCR). In clinical specimens, mVseq showed substantial agreement with single-target qPCR (92%; k statistic, 0.77; 259 of 282 viral tests); however, clinical sensitivity was reduced (81%), ranging from 62% to 100% for specific viruses. In 12 of the 47 patients tested, mVseq identified previously unknown BK virus, human herpesvirus-7, and Epstein-Barr virus infections that were confirmed by qPCR. CONCLUSIONS: Our results reveal factors that can influence clinical sensitivity, such as high levels of host DNA background and loss of detection in coinfections when 1 virus was at much higher concentration than the others. The mVseq assay is flexible and scalable to incorporate RNA viruses, emerging viruses of interest, and other pathogens important in transplant recipients.