摘要目的:探讨宏基因二代测序技术检测尿路感染患者尿液中多瘤病毒的临床价值。方法:2018年9月至2020年4月复旦大学附属中山医院收治的送检尿液宏基因二代测序的尿路感染患者共33例。采用随机扩增法、脱氧核糖核酸纳米球、华大基因公司BGI-500测序平台及BWA算法为基础的比对法等技术对尿液标本的核酸提取物进行二代测序分析，以尿标本中多瘤病毒核酸序列数>100为入选标准，分析检出多瘤病毒的类型、深度、覆盖度、丰度及序列数等参数。采用Taq Man探针实时定量聚合酶链反应（PCR）法验证并检测多瘤病毒DNA载量。结果:30.3%（10/33）的患者尿标本中检出JC多瘤病毒，其中1例合并检出BK多瘤病毒。所有患者在检测前均使用抗菌药物治疗，但仅一例尿培养阳性。JC多瘤病毒检出的深度均值为35.9，覆盖度均值96.8%，相对丰度均值92.6%，唯一比对序列数均值为3 154。实时定量PCR法检测10例患者尿中JC多瘤病毒DNA载量在5.84×10 3~6.55×10 8拷贝/ml，均值为6.82×10 7拷贝/ml，其中9例患者（均为免疫受损者）的病毒载量超过5×10 4拷贝/ml，与宏基因二代测序结果基本相符。每1 M数据中的严格比对序列数（RPM）与CD4 +T细胞计数存在线性负相关（ R2=0.84， P=0.004）。其中1例患者经西多福韦治疗后症状及尿检指标好转，随访尿液宏基因二代测序检查JC多瘤病毒的序列数及RPM值明显下降，实时定量PCR法证实尿中病毒DNA载量亦下降。 结论:宏基因二代测序技术在病毒性尿路感染的诊断中有着重要意义。西多福韦可能是治疗多瘤病毒尿路感染的一个选择。

abstractsObjective:To analyze the clinical value of polyomavirus detection in the urine by metagenomic next generation sequencing (mNGS).Methods:Urine specimens were collected from 33 patients diagnosed with urinary tract infection in Zhongshan Hospital of Fudan University from September 2018 to April 2020. The random amplification, DNA nanoball, BGI-500 sequencing platform, and Burrows-Wheeler Aligner algorithm were used for next generation sequencing and analysis of nucleic acid extracts from urine. The number of virus sequences in urine samples>100 was included in the study, and indexes of virus types, depth, coverage rate, abundance and reads number were observed. Viral load of polyomavirus DNA was determined using a TaqMan PCR strategy.Results:JC polyomavirus was detected in 30.3% (10/33) urine samples, with BK polyomavirus co-detected in one of them. All of the 10 patients were treated with antibiotics before, but only one with positive urine culture. The mean value of the depth, coverage rate, relative abundance and stringently mapped reads number (SMRN) were 35.9, 96.8%, 92.6% and 3 154, respectively. Further we verified the results with TaqMan real-time PCR, the viral DNA loads of the 10 patients were between 5.84×10 3 to 6.55×10 8 copies/ml, with mean value of 6.82×10 7copies/ml, 9 (all were immunocompromised patients) of them were higher than 5×10 4 copies/ml. The results were basically consistent with mNGS. There was a negative linear correlation between the RPM parameter (SMRN per M sequences) and the number of CD4 +T cells ( R2=0.84, P=0.004). In one case, after treatment with cidofovir, the hematuria and urine test indexes were improved, as well as the JC polyomavirus load tested by mNGS and RT-PCR. Conclusion:The clinical significance of polyomavirus detection in urine remains elusive, but the possibility of infection should be considered in cases of immunocompromised hosts. The mNGS is of great value in the diagnosis of viral urinary tract infection. Cidofovir may be effective in the treatment of urinary tract polyomavirus infection.