Tissue polymerase chain reaction for the diagnosis of cytomegalovirus disease after allogeneic hematopoietic cell transplantation
Year: 2016
Type: article
Study design: Observational Study
Abstract: Cytomegalovirus (CMV) infection remains a major complication of allogeneic hematopoietic cell transplantation (HCT) 1, 2. However, establishing the diagnosis of tissue-invasive CMV disease is not trivial due to the following reasons: (i) poor correlation between presence/quantitative level of CMV DNAemia/antigenemia and CMV disease 3, (ii) invasiveness of tissue-directed diagnostic procedures, (iii) viral shedding without true tissue involvement, and (iv) frequent coexistence of graft-versus-host disease (GVHD) 4, which can alter and/or mimic the classic morphology of CMV disease. Currently, diagnosis is based on (i) CMV DNAemia/antigenemia, (ii) morphological findings of CMV disease on hematoxylin and eosin (H&E)-stained specimens, and (iii) immunohistochemistry (IHC) using CMV antibody 5. However, a conclusive diagnostic modality when H&E- and/or IHC-based findings are equivocal or inconsistent is not available and toxic anti-CMV therapy can be detrimental in equivocal cases without tissue-invasive disease. For the first time, we show that tissue CMV PCR is a valuable adjunct to H&E/IHC in the diagnosis of CMV disease. Formalin-fixed paraffin-embedded (FFPE) specimens from gut or lung biopsy of 151 adult patients who underwent an allogeneic HCT at our center between 2010 and 2015 were retrieved from pathology archives. The availability of PCR results on the blood within 7 days of biopsy was required for inclusion. Patients received prophylaxis using acyclovir or valacyclovir at least for the period of iatrogenic immunosuppression. CMV surveillance was performed weekly or biweekly until Day +100. Endoscopy and bronchoscopy were at the physicians’ discretion. Surveillance for viremia was performed (since 2014) using the commercial Qiagen/Artus-Rotorgene quantitative real-time PCR assay with lower limit of detection of 100 IU/mL. Before 2014, the assay was based on quantitative real-time PCR based on a modification of an assay developed in the St. Louis Children's Hospital Virology Laboratory with lower limit of detection of 200 copies/mL 6. For DNA extraction, 3-5 FFPE tissue scrolls of 5–10 µm thickness were used. The procedure utilized the Qiagen EZ1 DNA Tissue Kit (Qiagen, Valencia, CA). Buffer G2 (190 µL) was added to the samples and then incubated at 75°C for 5 min. Next, 10 µL of proteinase K was added, and the samples were incubated at 56°C overnight. Approximately, 190 µL of each lysate was removed to a fresh tube and extracted on the EZ1 workstation using the EZ1 Paraffin Card. The purified DNA was eluted into a final volume of 200 µL. The PCR assay was a laboratory-developed real-time TaqMan assay using primers and probe targeting the CMV polymerase gene 6. The specimens were reviewed for H&E and IHC findings by an experienced pathologist blinded to clinical and molecular results. The result of each test was reported as positive, negative, or equivocal. Tissue PCR was performed by a laboratory technician blinded to clinical, H&E, and IHC findings. PCR results were reported as cycle threshold (Ct), defined as the number of cycles needed to detect a real signal above background fluorescence. Each specimen was classified into one of the three groups according to H&E/IHC results: (i) Double-positive, (ii) Double-negative, (iii) Equivocal. Group (iii) included any case that did not belong to groups (i) and (ii), ie those with equivocal results on one or both tests and those with a positive result on one test and negative result on the other. Groups (i) and (ii), collectively referred to as “concordant,” were used to determine the optimal Ct cutoff that maximized the agreement between tissue PCR results and double positivity/negativity. Biopsies were obtained from colon, duodenum, stomach, esophagus, and lung in 59 (39%), 44 (29%), 37 (24%), 7 (5%), and 4 (3%) cases, respectively. One hundred twenty-two (80%) cases were concordant (105 double-negative and 17 double-positive) and the remainder were equivocal. Viremic cases were marginally significantly more likely than non-viremic cases to be equivocal (28% vs. 14%, respectively; P = 0.055). Pathological evidence of GVHD was present in 57% vs. 52% of concordant vs. equivocal cases, respectively (P = 0.68). Similarly, these groups were not different with regards to the presence of clinical evidence of GVHD (82% vs. 90%, respectively). The optimal PCR Ct value for classification of concordant cases was 40, with good overall performance (AUC 0.91, P < 0.001), sensitivity 94%, specificity 79%, positive predictive value (PPV) 42%, and negative predictive value (NPV) 99%. Using this cutoff, 45% of equivocal cases were classified as negative. Table 1 shows the distribution of patients across H&E/IHC and PCR subgroups. Among viremic H&E/IHC-concordant cases, tissue PCR had a sensitivity of 100%, specificity of 50%, PPV of 44%, and NPV of 100%. Among non-viremic H&E/IHC-concordant cases, these numbers were 80%, 91%, 36%, and 99%, respectively. In this analysis on viremic and non-viremic cases, 31% and 62% of equivocal cases were classified as negative, respectively. One of the challenges in the diagnosis of CMV disease is H&E/IHC-equivocal cases. Avoiding potentially toxic anti-CMV treatment can be beneficial especially in non-viremic patients with negative tissue PCR. We demonstrate that a negative tissue PCR can be used to rule out CMV disease in H&E/IHC-equivocal cases. Although minimal spatial heterogeneity in the CMV target organ is assumed, we cannot rule out this possibility. Furthermore, we performed tissue PCR on FFPE specimens. The results may change if PCR were on fresh tissue, as it would be in the real-life setting. We expect the sensitivity of fresh tissue PCR to increase and specificity to decline. Until replicated in a controlled prospective study using fresh tissue samples, our results should not be used to guide treatment decisions. Our goal in this innovative, but preliminary, study was to assess the value of tissue PCR as an adjunct to H&E/IHC in the diagnosis of CMV disease, especially in equivocal cases. Future research could focus on clinical outcomes of non-viremic equivocal cases and whether anti-CMV treatment can be safely withheld in this group of patients. We could not address this question in the present work because all of our equivocal cases received anti-CMV treatment. Given the possibility of sampling error when using PCR on tissue biopsies, applying the same technique to fluid samples such as bronchioalveolar lavage or stool samples may be of value in future research. This study was funded by the Washington University Institute of Clinical and Translational Sciences grant UL1 TR000448 from the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH). AR, JFD, and GAS designed the study. KRV performed H&E and IHC. RSB and KMW performed PCR. AR wrote the manuscript. All authors critically reviewed the manuscript and approved the final draft Armin Rashidi1, Kiran R. Vij1,2, Richard S. Buller3, Kristine M. Wylie3,4, Gregory A. Storch3, John F. DiPersio1 1Bone Marrow Transplantation and Leukemia Section, Division of Oncology, Washington University School of Medicine, St. Louis, Missouri; 2Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri 3Department of Pediatrics, Washington University School of Medicine, St. Louis, Missouri 4The McDonnell Genome Institute, Washington University School of Medicine, St. Louis, Missouri Current address: Division of Hematology, Oncology, and Transplantation, Department of Medicine, University of Minnesota, 14-152 Phillips Wangensteen Building, 516 Delaware Street SE, Minneapolis, MN, 55455
Source: American Journal of Hematology
DOI: 10.1002/ajh.24609
Institution: Washington University in St. Louis
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2554101633