Background
Following solid-organ transplantation (SOT), the optimal dose of primary (val)ganciclovir (v(gcv)) prophylaxis against CMV infection is debated, and breakthrough infection and treatment-limiting side effects are frequently seen. Rates and risk factors for CMV prophylaxis breakthrough and premature cessation of prophylaxis for other reasons were investigated in a large cohort of consecutive SOTs.

Methods
SOT recipients transplanted (tx) between 2012 and 2016 at Rigshospitalet, and who were initiated on primary prophylaxis ≤14 days post-tx were followed from this time until 90 (±7) days post-tx. A prophylaxis score for each patient/day was calculated during the follow-up time (FUT) (score of 100 corresponding to the manufacturers’ recommended dose for a given eGFR; Figure 1). Prophylaxis breakthrough was defined as PCR verified CMV DNA positivity in plasma or BAL (i.e., infection) and premature stop of prophylaxis as >7 days with a score of 0. Time to event and hazard ratios (HR) were estimated with Cox models after adjustment for relevant risk factors.

Results
Of 585 SOTs (311 kidney, 117 liver, 106 lung, 51 heart) included, 41 (7%, 95% CI 4.9–9.1%) experienced CMV prophylaxis breakthrough (9/41 [22%, 9.2–34.6%] developed viral resistance to (v)gcv) and 33/585 (5.6%, 3.7–7.5%) ceased prophylaxis for other reasons during the first 90 days after tx. After adjustment for tx type, CMV IgG D+/R− mismatch and increasing % of FUT with a prophylaxis score <90 were associated with increased risk of breakthrough (HR 4.76 [95% CI 2.38–9.54] P < 0.001 and HR 1.15 [1.04–1.27] P = 0.007/10% longer FUT w/ score < 90 respectively, Figure 2) whereas tx type was not. The main risk factor for stopping prophylaxis for reasons other than breakthrough was lung tx (22.9%, HR 13.4 (vs. kidney SOT) [5.4–33.4]), mainly due to liver or myelotoxicity.

Conclusion
SOTs receiving (v)gcv primary prophylaxis doses below the manufacturers’ recommended doses according to latest eGFR were at an increased risk of CMV prophylaxis breakthrough, particularly in case of CMV IgG D+/R− mismatch, while 23% of lung tx recipients stopped prophylaxis mainly due to toxicity. Our findings indicate the need to dose adjust (v)gcv according to latest eGFR and preferably use novel, less toxic agents