Defining pp65 Enemia and qPCR Cut-Offs for Pre-Emptive Therapy of CMV Disease in Low-Risk, Seropositive Renal Transplanted Recipients

Abstract

CMV disease mostly invasive gastro-intestinal with low viremia occurs in approximately 16% of CMV seropositive renal transplant recipients not receiving Thymoglobuline (ATG). Pre-emptive therapy (PETh) to avoid disease in this group should cover the majority of patients at risk (high Negative Predictive Value – NPV). To define cut-offs for whole blood real-time quantitative PCR (qPCR) and antigenemia (pp65) assays and use PETh for CMV disease we collected blood samples weekly from day 7 to 120 after TX in pts CMV IgG+/ IgG+ donors who did not receive ATG. Results remained blinded. A suspicion of CMV disease required new qPCR/pp65 or biopsy. The highest value of pp65 and qPCR in pts without CMV disease one week before disease were used in ROC curves. As for Nov 2012, 92 pts had completed the collections or developed disease. They were males (54%), caucasians (69%), mean age 45±14y, 61% deceased donors, all under TAC/MPA/ Prednisone. Overall 12 pts (13%) had CMV disease (11 invasive gastro-intestinal/ 1 syndrome), at 64±21 days (31-98), and 80 did not. Of these 80 pts, 45 (56%) presented at least one episode of asymptomatic viremia and cleared it spontaneously (8 (18%) had PCR+/pp65+; 26 (56%) Pp65+/qPCR- and 11 (24%) qPCR+/Pp65-). Among pts with viremia or disease there was a trend for a higher pp65+ cells for disease than for viremia (152±352vs31±131/10 6 cells, p=0.073,) detected earlier for disease than viremia (64±21vs80±25 days, p=0.049). ROC curves (area 0.902±0.036, p=0.0001) revealed that pp65 of 4 cells had 83% sensitivity and 83% specificity to predict CMV disease. Using this cut-off, the NPV was 97%. The CMV copies detected by qPCR was higher for disease than for viremia (46657±50671 vs 5438±16681 copies, p=0.001) but time for detection was similar for both (74±25 vs 59±18 days, p=NS). ROC curve (area 0.903±0.056,p=0.0001) revealed that qPCR=844 copies had 83% sensitivity and 86% specificity to predict disease with the same NPV of 97%. These data indicate that both methods are adequate to detect CMV disease prior to its development in this low-risk population. Blood samples collection should start at day 30 and then weekly until day 120. When these cut-offs are reached, starting pre-emptive therapy would lead untreated only 3% of patients who would develop CMV disease.