CMV prevention strategies in allogeneic hematopoietic cell transplantation; the role of prophylaxis and pre‐emptive monitoring in the era of letermovir

The preferred strategy for preventing CMV in at‐risk populations in alloHCT has undergone a significant practice shift in recent years where the pendulum has swung from a pre‐emptive approach to now offering letermovir prophylaxis to all CMV seropositive recipients. Letermovir prophylaxis has resulted in significant reductions in post‐transplant clinically significant CMV infection (csCMVi) as well as other important outcomes such as CMV disease, resistant, and refractory CMV infections and nonrelapse mortality. However, prophylactic strategies are not without some limitations, namely delayed onset CMV infections, delayed CMV‐specific T cell immune reconstitution, increased drug costs and limited data within pediatric populations. Thus, this review aims to provide an overview of prophylaxis and pre‐emptive CMV preventative strategies, and how they are applicable in the current era of letermovir prophylaxis.

guide duration of prophylaxis.Other strategies for CMV prevention such as use of prophylactic infusion of third party multiviral-specific T-cells are under evaluation.Thus, this review aims to provide an overview of prophylaxis and pre-emptive CMV preventative strategies and how they are applicable in the current era of letermovir prophylaxis.

LETERMOVIR PROPHYLAXIS AS STANDARD OF CARE
In the past, CMV prophylactic studies in the alloHCT population were largely hindered by the availability of an anti-CMV-specific drug which was safe, lacked toxicity, and was efficacious. 6,7However, the landmark study by Marty et al. showing significant reduction in clinically significant CMV infection (csCMVi) (37.5% vs. 60.6%,p < .001)and allcause mortality by week 24 post-HCT (10.2% vs. 15.9%,p = .03)using letermovir prophylaxis in the first 100 days compared to pre-emptive therapy 4 was practice changing, leading to widespread adoption of letermovir prophylaxis and the highest level of recommendation in international guidelines. 3As a result, many institutes are now using letermovir prophylaxis in the first 100 days, although actual time of commencement varies across centres and patient CMV risk. 80][11] In a systematic review and meta-analysis of 48 studies in 7104 alloHCT recipients, letermovir prophylaxis was associated with a lower risk of CMV reactivation (OR 0.13, p < .05),lower csCMVi (OR 0.09, p < .05),lower CMV disease (OR 0.31, p < .05),lower all-cause mortality (OR 0.73, p < .01),and lower nonrelapse mortality (OR 0.65, p = .01). 5 Additionally, in a cohort of alloHSCT recipients on letermovir prophylaxis, a significant reduction of resistant and refractory CMV infection was observed (HR 0.15, 95% CI: 0.04-0.52). 1 The letermovir and preceding brincidofovir prophylaxis studies were also pivotal for introducing the clinical trial primary end point of clinically significant CMV infection (csCMVi), defined as any threshold of CMV DNAemia that triggers initiation of preemptive antiviral therapy at the discretion of the clinician or development of CMV disease. 7The definition takes into consideration individual patient risk factors, viral load kinetics, and institutional practices on deciding whether or not to commence pre-emptive treatment.Furthermore, csCMVi represents a conceptual shift away from starting antivirals upon reaching a specific viral load threshold given the longstanding difficulties of standardizing viral load diagnostics across different assay platforms as outlined by the World Health Organization. 12,13inical practice guidelines offer a recommended threshold viral load for low-risk patients of 280-900 IU/ml and high-risk patients of 137 IU/ml to commence pre-emptive treatment taking into consideration other patient and institutional factors. 3Use of csCMVi has now become an end point well accepted by relevant regulatory agencies and allows for consistency in clinical research as well as trials moving forwards. 4,14

Delayed-onset clinically significant CMV infections
A limitation of any prophylaxis strategy, which aims to prevent infection during the highest risk period, is rebound CMV infection following cessation of prophylaxis due to delayed CMV reconstitution. 46][17][18][19] Risk factors for delayed csCMVi were either due to baseline transplant-related factors or complications arising after alloHCT such as graft versus host disease (GVHD) and need for immunosuppression beyond the high-risk period.Mismatched grafts, haploidentical donor transplants, umbilical blood cord source, T cell depletion, CMV R+/D-serostatus, development of GVHD, and concomitant use of steroids have also been identified as risk factors for delayed csCMVi. 16,18In retrospective studies, the highest risk for delayed csCMVi was conferred by mismatched donor (HR 13.0, p = .011)and to a lesser extent CMV R+/D-transplants (HR 2.39, p = .043). 18V seropositive recipients receiving umbilical blood stem cells are considered one of the highest risk groups for CMV infection and disease.20,21 Extended letermovir prophylaxis in this population from 100 days to 6 months was associated with a very significant reduction in csCMVi compared to a historical control (0% vs. 82%, respectively).20 In another cohort of umbilical cord transplant recipients, the cumulative incidence of csCMVi at the end of letermovir prophylaxis compared to high dose valaciclovir was significantly lower (19% vs. 65%, p < .001).22 However upon letermovir discontinuation, the development of csCMVi was high compared to controls receiving high dose valaciclovir so that by 12 months, both groups had similar rates of csCMVi (60% vs. 65%, p = .2). 22 As a result, several studies have evaluated whether extending prophylaxis duration would be more effective in patients who remained at high risk for CMV.11,14,23 In a retrospective study comparing primary prophylaxis for greater than 14 weeks (median 116 days) compared to a historical control of pre-emptive monitoring, high-risk haploidentical and T cell depleted grafts had significantly reduced csCMVi (7.1% vs. 68.3%,p = .0003 respetively) by 6 months.23 A randomized international multicentre study of 220 patients with ≥1 risk factor for csCMVi (related or unrelated donor with ≥1 mismatch, haploidentical donor, cord blood source, ex-vivo T cell depleted graft, use of anti-thymocyte globulin [ATG] or alemtuzumab, GHVD or use of prednisolone ≥1 mg/kg) to either extended letermovir prophylaxis from day 100 to 200 or matching placebo.14 Initial results demonstrated that the extended letermovir arm had lower incidence of csCMVi (1.4% vs. 17.6%, p = .0005)compared to placebo from week 14 to week 28 meeting the primary endpoint.14 However, upon discontinuation of letermovir, delayed csCMVi beyond week 28 was once again observed with similar incidence of csCMVi between week 14 and 48, between the extended letermovir arm versus placebo (14.6% vs. 20.3%,p = .16). 14 Importantly, no safety concerns were identified and all-cause mortality between both arms were similar.14 The study showed that reduction of csCMVi is effective while on letermovir prophylaxis however delayed or rebound csCMVi may still occur and is not altered by the duration of letermovir. 14A possible explanation for this observation is that prophylaxis diminishes CMV antigen exposure thus delaying the development of recipient CMVspecific T cell reconstitution required for effective and sustained viral control.24

Delayed T cell immune reconstitution-a role for CMV immune monitoring?
To determine the impact of letermovir prophylaxis on the development of CMV-specific immunity after alloHCT, Zamora et al.
prospectively compared 56 letermovir recipients compared to 93 pre-emptively treated recipients to evaluate CMV-specific T cell immunity. 24At 3 months, polyfunctional CD4+ and CD8+ CMVspecific T cell responses to CMV antigens IE-1 and pp65 in letermovir recipients were significantly lower than controls. 24Of the letermovir patients who were exposed to CMV DNAemia and had viral shedding, increased CMV T-cell responses were observed 24 suggesting that the presence of virus may lead to a more robust CMV T-cell reconstitution.
To expand on that concept, studies have been performed evaluating the utility of commercially available CMV immune monitoring assays and whether they can identify transplant recipients who have developed CMV immunity by the end of prophylaxis. 25,26CMV immune monitoring in the solid organ transplant population has shown clinical utility by identifying patients who have developed CMV immunity and in whom CMV prophylaxis could be discontinued. 25,268][29][30] Most recently, a large multicentre study of 241 CMV R+ alloHCT patients who underwent a CMV ELISPOT assay at different time points after HCT showed that patients with high cell mediated immunity (CMI) were less likely than those with low CMI responses to develop csCMVi p < .001. 29 In addition, high IE-1 responses seemed to be protective and further analysis showed that a decrease in IE-1 T cell responses between baseline and week 4 of transplant was associated with increased all-cause mortality (log rank p = .04). 31 Monitoring using the QuantiFERON-CMV assay in 94 alloHSCT patients showed that a positive/reactive result was associated with a lower peak CMV viral load 2 while low functional T cell responses as measured by interferon-gamma IU/ml to nonspecific mitogen was associated with increased non-relapse mortality. 32rther studies are required to assess whether CMV immune assays can be routinely incorporated into CMV prevention strategies including letermovir prophylaxis and how they could guide clinical decisions on duration of prophylaxis.

Identifying highest R+ risk groups-targeted letermovir use
Not all CMV seropositive recipients are at the same or equal risk for csCMVi and thus some assessment of the CMV risk profile is required to ensure that the highest risk group are appropriately targeted and where low risk groups could avoid unnecessary letermovir use.Older ganciclovir prophylaxis studies aimed at reducing CMV disease estimated that 60%-65% of recipients may unnecessarily be prescribed prophyalxis. 33Better selection of the high-risk group was addressed in a study, which developed a risk score for low, intermediate-low, intermediate-high, and high-risk category using six predictors; donor CMV serostatus, type of donor, conditioning intensity, total body irradiation, ATG and mycophenolate use. 34Applying this risk score in a prospective cohort, those in the low risk group (n = 63) were not prescribed letermovir. 35Although 18 of those patients (28.6%) were later commenced on letermovir (at median 28 days) with no patient developing subsequent csCMVi, few patients in the remaining 45 patients developed csCMVi (11%) and were safely treated with a single course of pre-emptive therapy without complication. 35This study demonstrates a risk-based strategy for prophylaxis allowing sparing of some low risk patients from receiving letermovir. 35A randomised study evaluating this strategy would be important to perform.

Secondary prophylaxis-the role of letermovir
Understanding the efficacy of letermovir as secondary prophylaxis in patients who have achieved viral clearance with pre-emptive therapy but require sustained viral suppression due to continued immunosuppression has been evaluated in several studies. 23,36,37In 80 consecutive CMV R+ patients who received letermovir as secondary prophylaxis, only 5.5% developed breakthough csCMVi 36 while in a smaller cohort of 14 patients, no patient experienced csCMVi, 23 demonstrating good efficacy when used in this setting.An open label single arm study in high-risk HCT patients using letermovir secondary prophylaxis is ongoing. 37Importantly, letermovir should only be commenced once viral suppression has been achieved due to the low barrier of developing letermovir resistance. 38,39

Cost and equitable global access to letermovir
Despite the evidence and recommendation to use letermovir prophylaxis in CMV seropositive alloHSCTs as an effective CMV prevention strategy, not all countries and institutes have financial access or reimbursement processes to use the drug or to provide it in an equitable manner for all patients. 40,41A recent study of HCT recipients of selected countries outside of Europe and North America, found that pre-emptive therapy was the preferred approach to manage CMV and no centres reported on the use of letermovir. 40Unfortunately, the rates of csCMVi, recurrent CMV, resistant and refractory CMV, CMV disease and CMV-related mortality were reported to be high. 40Additionally, several countries have reduced or no access to second line anti-CMV antivirals such as foscarnet and cidofovir for treatment of resistant and refractory CMV. 40Access to recommended and effective CMV antivirals is of utmost importance to ensure the best outcomes possible. 42In the absence of letermovir, high dose oral valaciclovir (6-8 g daily) has been shown to be a moderately effective prophylactic strategy in high-risk haploidentical, T-cell depleted CMV seropositive patients by reducing the incidence and time of onset of csCMVi (HR 0.32, p = .0005). 41Similar findings have previously been demonstrated in umbilical blood cord transplants. 21,43Thus high dose valaciclovir, may be considered as an alternative prophylactic strategy if letermovir is unavailable.
Cost-effectiveness studies performed in different settings and countries have found letermovir prophylaxis to be cost effective compared to pre-emptive treatment alone. 44,45In the US setting, letermovir yielded an ICER of US$59,356 per QALY gained. 44

Pediatric population
A phase 2 single arm, dose-finding, study of letermovir for CMV prevention in pediatric patients has been recently completed (Clinicaltrials.govNCT05711667).So far results are reported only for the age group 12-17 years of age.7][48][49] In the absence of letermovir, primary prophylaxis using valganciclovir, ganciclovir, or foscarnet may be considered in high risk pediatric patients, although recommendations for specific high risk groups are yet to be determined. 3,50

Other prophylaxis therapies-3rd Party and donor derived CMV T-cells
Another exploratory strategy aimed at the prevention and treatment of CMV in alloHCT recipients is the infusion of virus-specific adoptive 3rd party or donor derived T-cells; either CMV specific or multivirus specific.Infusion and restoration of CMV-specific T cells during the high-risk period until host CMV-specific immunity is restored may prevent the need for prolonged duration of antivirals and provide a sustained immune response.Previous single arm studies have shown that infusion of donor derived CMV-specific T cells at median day 45 following alloHCT were safe and reduced csCMVi (17% vs. 36% p = .01)and the total number of antiviral treatment days in those receiving CTLs without a reduction in CMV infection rates was 3.4 versus 8.9 days, p = .03. 51 Multiviral T cells against 6 dsDNA viruses (CMV, EBV, HHV6, AdV, BK, JCV) have been evaluated in Phase 2 study as a preventative therapy and were well tolerated with minimal toxicity and no clinically significant viral infections observed by week 14. 52Randomized trials are required and are currently underway (Clinical trials.govNCT05305040).

3.8
The role of pre-emptive monitoring Preemptive monitoring and prophylaxis are complimentary strategies and should be in place at all centers.While the need for routine CMV monitoring during letermovir prophylaxis has not been formally evaluated, the majority of centers currently continue routine monitoring for all patients at risk for CMV regardless of prophylaxis.Single low level CMV loads "blips" may be observed during letermovir prophylaxis which can represent abortive or non-infectious DNA rather than replicating virus requiring treatment. 53Rebound CMV infections after letermovir prophylaxis typically occur during the first 4-6 weeks after stopping letermovir. 54Routine monitoring and preemptive therapy is the recommended approach for low risk patients such as such as CMV seronegative recipients receiving seropositive grafts who are not candidates for letermovir.We recommend routine monitoring after discontinuing letermovir for as long as patients remain at risk for CMV to inform need for preemptive therapy.Furthermore, as letermovir is mainly administered orally, those patients who develop intolerance, are unable to swallow tablets or develop multi-organ failure necessitating the early cessation of letermovir, should continue pre-emptive monitoring.It also remains to be seen whether some CMV R+ populations could be identified to be monitored as part of a pre-emptive treatment strategy in order to benefit from low level CMV antigen exposure to facilitate host CMV immune restoration as observed in liver transplant recipients, albeit with CMV disease as the outcome. 55

CONCLUSION
The clinical area of CMV prevention and risk stratification in the adult alloHCT population has undergone significant and exciting practice changes in the last few years since letermovir prophylaxis was found to be effective and made widely available.Real world data have convincingly demonstrated that compared to pre-emptive treatment, prophylaxis reduces not only csCMVi and CMV disease but also reduces resistant and refractory CMV infection, nonrelapse mortality and all cause mortality.However, there are still further refinements and improvements to be made to prophylaxis strategies regarding better risk identification, facilitating CMV immunity and optimal prophylactic duration.Pre-emptive monitoring and treatment continues to have a role in patients who are at lower risk of CMV or unable to be prescribed letermovir.

ACKNOWLEDGMENTS
Open access publishing facilitated by The University of Melbourne, as part of the Wiley -The University of Melbourne agreement via the Council of Australian University Librarians.