This retrospective study evaluated the prophylactic efficacy of LET against CMV reactivation and the transplant outcomes after aHSCT in clinical practice. The findings demonstrated a definite prophylactic anti-CMV effect of LET, along with an excellent safety profile. Although LET prophylaxis did not significantly improve NRM, LET may have improved OS by inhibiting and delaying csCMV infection.
Preemptive anti-CMV treatment, guided by the monitoring of CMV reactivation after aHSCT, has been confirmed to be a successful approach for CMV disease prophylaxis over the last few decades. This strategy was facilitated by the advent of effective anti-CMV agents, such as ganciclovir, foscarnet, and valganciclovir, and with the introduction of sensitive CMV monitoring methods. Nevertheless, seropositivity for CMV remains a risk factor for NRM4,12,13, and CMV reactivation early after aHSCT is associated with increased NRM3,4. The contribution of CMV reactivation to morbidity and mortality after HSCT is considered to be attributable to indirect effects. CMV shows immunomodulatory activity; for example, it can inhibit interferon-γ production and its receptor signal and promote the production of IL-10 in infected host cells. Additionally, CMV can prevent antigen presentation by its peptides and NK cell function. Therefore, CMV reactivation can cause complications associated with the host's immune conditions through these modifications.
Regarding the prophylaxis for csCMV infection, the cumulative incidence at 180 days after aHSCT in the present study was slightly higher than that reported in the phase 3 study. However, compared with the phase 3 study, this investigation included a higher percentage of CMV high-risk patients, such as those who received cord blood transplantation (CBT) and HLA-mismatched hematopoietic stem cell transplantation (HSCT). CBT has been reported to be a risk factor for CMV reactivation or disease14,15, primarily as a result of the immaturity and dysfunction of lymphocytes in the cord blood16,17, and HLA mismatch is also thought to contribute to CMV reactivation18,19. T cell-replete aHSCT from an HLA haploidentical donor with post-transplant cyclophosphamide (PTCy) has also been associated with a high risk of CMV20,21. A recent report revealed that PTCy increased the risk of CMV infection in CMV-seropositive recipients who received aHSCT from HLA-haploidentical/matched sibling donors compared to recipients who received aHSCT from HLA-matched sibling donors with conventional calcineurin inhibitor-based GVHD prophylaxis. This suggests that PTCy itself may increase the risk of CMV infection in CMV-seropositive recipients regardless of HLA disparity22. Alternatively, Terao T et al. reported that LET prophylaxis may be associated with an increased risk of chronic GVHD through an early increase in HLA-DR+ activated T cells in PTCy-haplorecipients who received LET23. The ECIL 7 recommends real-time quantitative PCR (qPCR) methods to guide the initiation of preemptive therapy and to monitor the response11. We used the anti-CMV pp65 monoclonal antibody HRP-C7 assay to evaluate CMV reactivation. Kanda et al. reported that CMV antigenemia and qPCR were equally appropriate as CMV monitoring techniques13. Nevertheless, they demonstrated that the threshold for CMV antigenemia positive cells, usually used for CMV high-risk patients in Japan, may be too low. Therefore, for some Japanese patients, we may be initiating preemptive therapy excessively compared to patients monitored by qPCR.
Consistent with the results of the phase 3 study10, the onset of csCMV infection was significantly delayed by LET prophylaxis in the present study. Delayed CMV reactivation has been associated with a reduction in transplant-related mortality in the era of preemptive therapy24. Therefore, CMV prophylaxis with LET may reduce transplant-related mortality. In addition, CMV viral load is associated with an increased risk of early death24,25. According to the results of the phase 3 study, the median CMV DNA level with csCMV infection in the LET group (223 copies/mL [range: 150–27,946 copies/mL]) was lower than that in the control group (1,014 copies/mL [range: 150–106,830 copies/mL]), although the median CMV DNA level without csCMV infection was comparable in both groups10. All-cause mortality at week 24 after aHSCT was lower in the LET group than in the non-LET group, whereas the NRM of the LET group was slightly, but not significantly, lower than that in the non-LET group. In particular, the risk of all-cause mortality in patients with csCMV infection is reduced by LET prophylaxis26. Although LET prophylaxis was not associated with improved OS with statistical significance in this study, we believe that LET has the potential to improve transplant outcomes. A recent report from the adult T-cell leukemia/lymphoma (ATL) endemic area of Japan, in which the cumulative incidence of csCMV infections was considerably high, probably owing to the higher susceptibility of ATL patients to CMV infections27, suggested that LET reduces transplant-related mortality28. Therefore, further analysis with a large-scale survey is needed to clarify the effects of LET on transplant outcomes in clinical practice.
The optimal timing for initiation of LET prophylaxis remains unknown. Unlike the phase 3 study, in which LET administration was initiated at a median of 9 days following HSCT, all patients in this study received LET prophylaxis from the day of aHSCT. We believe that early initiation of LET prophylaxis after aHSCT may enhance the benefits for recipients. The phase 3 study was designed to initiate LET prophylaxis by day 28 after aHSCT10. Forty-eight patients already had detectable CMV DNA at the time of LET initiation, and the median day to randomization was longer than that of patients with undetectable CMV DNA (15 days vs. 8 days, P <0.001). The Kaplan–Meier event rate of csCMV infection (44.3%) was higher than that in patients with undetectable CMV DNA (18.9%)29. Additionally, pre-engraft CMV DNAemia developed in some patients, as reported previously30, although we could not evaluate CMV reactivation before engraftment because we used the CMV pp65 antigenemia assay to detect CMV reactivation. In a recent report from a German group31, the median starting time for LET prophylaxis was 10 days later than that in the phase 3 study. The cumulative incidence of csCMV infection through week 14 was significantly lower in the LET group than in the control group. However, unlike the phase 3 study and our present study, the median time to CMV reactivation in patients with csCMV was similar in both groups. These results emphasize the importance of early initiation of LET treatment.
This was a small-scale retrospective study, and the control group comprised patients who had undergone aHSCT before LET was approved (January 2011 to April 2018). However, transplant outcomes are generally improved by other factors as well, including new antibiotics and examination techniques32. Consequently, the transplant outcomes of patients who recently underwent aHSCT without LET prophylaxis might be better than the findings obtained in this study.
In conclusion, LET prophylaxis for csCMV infection was beneficial for aHSCT recipients in a real-world situation. However, a large-scale study is required to ascertain the appropriate timing for LET prophylaxis initiation and confirm the influence of LET on transplant outcomes.