LAUSR

We thank Dr. Frange and Dr. Leruez-Ville for their comments regarding our recent article (1). First, due to the inherent limitations of a case series, we agree that it is impossible to rule out potential confounding benefits from concurrent intravitreal treatments. While such confounding cannot be eliminated, the cases of successful CMV retinitis treatment with intravitreal dosing cited by Frange and Leruez-Ville involved either continuous delivery via ganciclovir implant or more than 4 serial injections (2–4). None of our patients received any drug-delivery implants (which are not currently available in the United States) and only one (patient D) received a comparable regimen to that described by Fan et al. (4). As presented in Fig. 1 of our recent article (1), patient A received just a single intravitreal foscarnet dose; patient C received none; and patient D received 3 total injections—fewer doses than described in prior case reports of successful treatment by intravitreal therapies alone. Second, Frange and Leruez-Ville emphasize the important fact that viral subpopulations may have different susceptibilities. The possibility exists that wild-type virus was present within the retinas of our patients, while resistant virus was isolated from blood. We confirmed this finding in one patient (patient B), in whom CMV resistance testing of aqueous fluid was performed. In this case, no UL97 or UL54 resistance mutations were identified in aqueous fluid, sent for testing prior to initiation of letermovir therapy. The potential for compartmentalization will be an important consideration for larger studies. Third, this case series was not designed to assess the mechanisms by which letermovir resistance emerged. We cited the prior in vitro work by Chou et al. suggesting that mutations conferring letermovir resistance appear rapidly in culture and agree that this mechanism is most likely (5). At the same time, since we did not conduct deep sequencing of viral isolates before and after receipt of letermovir, we could not make definitive conclusions about the manner by which letermovir resistance arose. As such, we felt it was prudent to present the potential for resistant CMV subpopulations as an alternative mechanism. Overall, we are in agreement that caution is warranted with use of letermovir for therapeutic rather than prophylactic or suppressive purposes. We are hopeful that future research will better address some of the interesting and important questions raised here.