Optimal Management Approach to Prevent Cytomegalovirus Infection in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

A 51-year-old man with acute myeloid leukemia (AML) and unfavorable-risk cytogenetics attained a first complete remission using conventional induction chemotherapy. He has an HLA-matched related donor (his brother) who agrees to donate hematopoietic cells for an allogeneic transplant. Both the patient and donor are cytomegalovirus (CMV) seropositive.

What is the optimal approach to prevent CMV infection in patients undergoing allogeneic hematopoietic cell transplantation (AHCT) in 2018?

Dating back to the earliest AHCT reports, CMV infection has been identified to be one of the most important complications. CMV infectious pneumonia, specifically, was the most feared obstacle due to its association with very high mortality. Although the unfavorable outcomes in treating established CMV disease certainly have decreased in recent decades, they still remain high, particularly if CMV pneumonia develops.¹  CMV-seropositive AHCT recipients have an overall higher rate of mortality than CMV-seronegative patients, especially in transplants from unrelated or mismatched donors; the risk is even greater if the donor is CMV-seronegative.^2,3  Furthermore, recipients have a higher nonrelapse mortality if CMV reactivation occurs less than 100 days after transplantation.⁴  Therefore, strategies to prevent the development of CMV disease should be used. Many different improvements in CMV preventive strategies have been introduced, including new, sensitive, rapid diagnostic tests; recognition of risk factors; and development of antiviral preventive strategies.

Nearly three decades ago, Dr. Joel D. Meyers and colleagues⁵  recognized that CMV viremia was an important risk factor for the development of CMV disease. As blood test results significantly influenced patient treatment, the concept of preemptive therapy based on a rapid diagnostic test was introduced. The first test fulfilling these requirements was the so-called “shell vial culture.” In a landmark study, Dr. James M. Goodrich and colleagues⁶  showed that by using this test as a guide to introduce antiviral therapy, it was possible to dramatically reduce the risk for clinically significant CMV disease and thus improve overall survival (OS). Today, most centers use quantitative polymerase chain reaction (PCR) on blood samples to monitor patients and guide the use of antiviral therapy. With this strategy, the risk for early CMV disease (developing within the first 100 days after AHCT) is now less than 3 percent, and death directly caused by CMV disease is rare at most centers, though disease occurring later after AHCT is not uncommon.

Despite these advances, several studies have shown that recipients who are CMV-seropositive remain at risk for worse OS and increased transplant-related mortality. Furthermore, developing a primary CMV infection in a pretransplant CMV-seronegative patient is also associated with poorer survival. The latter can be prevented by using “CMV-safe” blood products and, if possible, a CMV-seronegative donor. The importance of donor CMV serological status for a CMV-seropositive patient remains controversial.

Can the negative effect of being a CMV-seropositive transplant recipient be overcome? The logical solution would be to use an effective antiviral prophylaxis similar to the successful strategies for preventing herpes simplex virus and varicella-zoster virus. And is the negative effect then due to CMV replication itself or is there a different mechanism? Support for the theory that viral replication is bad for the patient was presented in an article by Dr. Margaret L. Green and colleagues⁷  using a retrospective cohort of patients for whom monitoring had been performed using plasma-based PCR. They found that CMV replication indeed was associated with an increased risk for overall and all-cause mortality independent of the use of preemptive therapy. Furthermore, the risks increased with increasing viral load. Thus, it would be logical to prevent CMV replication rather than wait for it to develop and then intervene with preemptive antiviral therapy.

What are the options for preventing CMV replication? Until now, the two most effective antiviral agents have been IV ganciclovir and foscarnet. The use of both drugs, however, has been hampered by the considerable adverse effects. In a prospective, randomized, placebo-controlled study, ganciclovir use reduced the risk for CMV disease but failed to improve OS and was associated with neutropenia and bacterial infections.⁸  As a result, ganciclovir prophylaxis therapy has been limited. No controlled study has been performed with foscarnet — an agent known to cause marked serum electrolyte abnormalities. Other well-studied prophylactic options include high-dose intravenous acyclovir, oral valaciclovir, and intravenous immune globulin; all agents have been regarded as having low efficacy.⁹  Furthermore, in a recent retrospective study examining the intensity of anti-CMV prophylaxis in haploidentical transplant recipients, less potent regimens were associated with earlier onset and an increased incidence of CMV reactivation.¹⁰  Therefore, interest has been directed toward the study of new antiviral drugs and anti-infective vaccines. Despite initially promising results in phase II studies, two new antiviral agents and one vaccine failed in the pivotal phase III studies. In a randomized, placebo-controlled study, maribavir failed to meet the primary endpoint of reducing CMV disease¹¹ ; this agent is being studied now in much higher dosages for the treatment of resistant or refractory CMV infections. An investigation using intravenous brincidofovir also was unsuccessful; the primary endpoint of lowering clinically significant CMV infection (CMV disease or the need for preemptive therapy) in a randomized, placebo-controlled study was not met, and treatment was associated with significant toxicity.¹²  Finally, recently released results of a randomized phase III study using the anti-CMV vaccine ASP0113 showed inability to meet the primary and secondary endpoints.¹³ 

On the other hand, letermovir is a highly specific antiviral agent against human CMV that possessed a unique mechanism of action by inhibiting the CMV terminase complex. Dr. Francisco M. Marty and colleagues¹⁴  recently reported the results from a randomized, placebo-controlled, phase III clinical trial conducted in CMV-seropositive AHCT recipients in which letermovir was administered through week 14 after HCT. A total of 495 patients had undetectable CMV DNA at randomization. Only 37.5 percent of patients (122 of 325) given letermovir prophylaxis had clinically significant CMV infection or a primary endpoint event by week 24 after AHCT, compared with 60.6 percent of the placebo group (103 of 170; p<0.001). Overall, the frequency and severity of adverse events in the two groups were similar. At 48 weeks after AHCT, all-cause mortality was 20.9 percent among letermovir recipients and 25.5 percent among placebo recipients. The drug was well tolerated also, and did not demonstrate toxic effects on the bone marrow, kidney, or liver. Both the U.S. Food and Drug Administration and the European Medicines Agency recently approved the drug based on these study results.

Will letermovir prophylaxis therapy now become the new standard of care? The study results discussed here certainly are promising, but the optimal fashion to use the drug will require additional studies. For example, it remains unclear if the 14-week duration of prophylaxis used in the study by Dr. Marty and colleagues is the optimal regimen. Although letermovir was superior to placebo in both high-risk and standard-risk patients, the results show an increased risk for developing significant CMV infection after cessation of prophylaxis in the high-risk group. Such data make the case that longer prophylaxis might be beneficial. Other unknowns include the likelihood for subjects to develop CMV resistance to letermovir, though the risk was low in this study.

A more important consideration perhaps, is whether the use of an efficacious antiviral agent will increase the risk of leukemia relapse. Several reports have shown, especially in AML patients, that leukemia relapse risk is lower in the setting of CMV viremia.^15-17  Other studies have not been able to demonstrate this association, including the large Center for International Blood and Marrow Transplant Registry study by Dr. Pierre Teira and colleagues.⁴  These investigators showed that CMV replication was associated with increased transplant-related mortality and decreased OS without showing a positive effect on relapse in any disease. Although the randomized letermovir investigation did not show an effect on risk of relapse, it should be recognized that the number of patients treated was too low and the follow-up too short to assess this risk. Through further investigation and more careful assessment of these considerations (and with more widespread use of letermovir and other effective anti-CMV agents), we might be able to answer this important question.

In addition to these issues, the economic burden of dealing with CMV infections is significant. Dr. Christine Robin and colleagues¹⁸  reported that the financial costs of AHCT increase by 25 to 30 percent in the setting of CMV episodes — a difference that could be reduced by appropriate use of prophylactic strategies. Given what we know today, letermovir seems to be an important development in the long-term struggle to reduce the negative effect of CMV on the outcomes of patients undergoing AHCT. However, letermovir has not been studied in children or in CMV-seronegative subjects.

Given the current approaches, we would use an antiviral prophylaxis strategy using letermovir in adult patients at high risk for CMV infection, while patients at lower risk and children could be treated using the current strategy of monitoring with quantitative PCR and preemptive antiviral therapy.