Virus-Specific T-Cell Therapy for Prophylaxis and Treatment of Cytomegalovirus Infections After Transplantation: a Scoping Review

Abstract

Background

Cytomegalovirus (CMV) infection is a leading complication following hematopoietic stem cell transplant (HSCT) and solid organ transplant (SOT). Virus-specific T cells (VSTs) have been used for the prophylaxis and treatment of CMV infections. We conducted a scoping review to catalogue and characterize the existing literature.

Methods

Systematic searches were performed in collaboration with an expert librarian. Inclusion criterion was the use of CMV-VST for prophylaxis or treatment in HSCT and SOT patients. Major exclusion criteria were case reports and series with <5 cases. Databases were queried from inception to 31 May 2024. Of the 2587 identified abstracts, full text review was performed on 92 articles, and 67 studies underwent final data extraction.

Results

Most studies were in the HSCT population. The CMV infection rate was 28% (interquartile range [IQR], 14%–44%) when CMV-VSTs were used as prophylaxis. Response rates for non–refractory and/or resistant (R/R) infections and R/R infections in HSCT patients were 98% (IQR, 70%–100%) and 70% (IQR, 56%–88%), respectively. Four studies included SOT patients with R/R infections, demonstrating a response rate of 15%–64%. Variables including donor/recipient serostatus and antiviral use were heterogeneously reported, and various definitions of CMV infection and response were used. CMV-VSTs were well-tolerated with minimal adverse events reported.

Conclusions

CMV-VSTs are more commonly used in HSCT patients with limited data in SOT patients, and differential reporting of key variables precludes extrapolation. A standardized registry should be considered for future studies with additional focus on the optimal dosing, timing, and interaction with concurrent antivirals.

Cytomegalovirus-specific T cells have been used in studies of primarily adult hematopoietic stem cell transplant recipients. Heterogenous trial design and data reporting make interpretation and clinical translation challenging. Standardization of study design and reporting are needed to address important knowledge gaps.

Graphical Abstract

Graphical Abstract.

This graphical abstract is also available at Tidbit: https://tidbitapp.io/tidbits/virus-specific-t-cell-therapy-for-prophylaxis-and-treatment-of-cytomegalovirus-infections-after-transplantation-a-scoping-review-4d91270d-0bfe-4952-8501-5b55fc46cb91?utm_campaign=tidbitlinkshare&utm_source=ITP

Cytomegalovirus (CMV) infection is a leading complication in patients following hematopoietic stem cell transplant (HSCT) and solid organ transplant (SOT) [1]. Risk factors include donor (D) and recipient (R) serostatus, induction and prophylactic immunosuppression regimens, antiviral prophylaxis, and underlying immune dysregulation [2, 3]. Once primary infection or secondary reactivation of CMV occurs, it may progress to CMV syndrome or end-organ disease. CMV also exerts indirect immunomodulatory effects and increases the risk of other opportunistic infections and graft failure [4].

Treatment options for CMV infection are limited. The use of first-line agents is often constrained by the emergence of myelosuppression [5]. Second- and third-line alternatives are associated with significant nephrotoxicity and treatment-emergent resistance or relapse [6].

Technological advances have accelerated the development of virus-specific T-cell (VST) therapeutics [7–9]. CMV-VSTs can be generated and expanded ex vivo using antigen-presenting cells loaded with peptides, viral vectors encoding immunodominant antigens such as pp65 and IE1, or by magnetic enrichment [10]. Depending on factors such as D/R serostatus and the availability of suitable seropositive related or unrelated donors, CMV-VSTs can be generated using autologous or allogeneic T cells; however, allogeneic sources are most feasible for clinical scaling [11]. Third-party donors are particularly important in D⁻/R⁺ patients due to the lack of donor CMV memory T cells for immunity. VSTs from third-party donors can also be cryopreserved and used as a “banked” product for multiple unrelated patients when necessary [12, 13].

The objective of this scoping review is to identify and collate published series and trials of CMV-specific VST use in HSCT and SOT recipients, to describe patient characteristics and clinical outcomes, and to inform future clinical trial design and reporting parameters.

METHODS

Search Strategy

The search strategy was formulated in collaboration with an expert medical librarian. Keywords representing the concepts “cytomegalovirus,” “T-cell therapy,” “organ transplantation,” and “stem cell transplant” were queried in PubMed and Embase. Four queries were performed up to and including 31 May 2024 (Supplementary Appendix 1). Results were exported to a review management database (Covidence, Veritas Health Innovation, Melbourne, Australia). The search yielded 2929 articles and 342 duplicates were removed (Figure 1). Registration of the protocol can be found under the Open Science Framework at https://osf.io/3cun9.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram of the systematic search.

Screening and Data Extraction

Two authors independently screened citations and completed data extraction (Supplementary Appendix 1). Titles and abstracts of 2587 articles were screened, and 92 underwent full text review; in total, 67 articles were included in data extraction. Studies met inclusion criteria if they evaluated the use of VSTs for CMV prophylaxis or treatment in pediatric or adult HSCT or SOT populations. Exclusion criteria were case reports and series with <5 cases, duplicates, technological protocols, language other than English, and animal studies.

Statistical Analysis

Descriptive analyses were used to summarize the studies. Continuous data were expressed as medians with interquartile range (IQR) and categorical variables were summarized as proportions. Analyses were stratified by VST indication: prophylaxis, treatment for nonrefractory infection, and treatment for refractory infections. All calculated values were rounded to the nearest whole number.

RESULTS

General

The search yielded 67 articles that proceeded to full text review, of which 18 were abstracts, 4 were letters to the editor, and 45 were complete publications. Study designs were predominantly phase 1/2 studies (33 studies), followed by case series (27 studies), cohort studies (4 studies), and randomized controlled studies (3 studies) (Tables 1–3). CMV-VSTs were used as prophylaxis in 22 studies, for treatment of nonrefractory infections in 15 studies, and for resistant/refractory (R/R) infections in 38 studies. In total, there were 1466 HSCT cases reported from 65 studies and 41 SOT cases from 4 studies. It was not possible to identify which cases represented unique patients, or those that were reported multiple times. Of the included studies, 27 reported both adult and pediatric cases, 16 reported adult only, and 9 pediatric only; 15 studies did not report patient age. Thirty-four studies did not report patient sex and in the 33 remaining studies, female patients comprised between 20%-67% of the included patients. Human leukocyte antigen (HLA) matching between transplant donor and recipient, and immunosuppression regimens used for conditioning and graft-versus-host disease (GVHD) prophylaxis, were heterogeneously reported across the studies (Supplementary Appendix 2).

Table 1.

Summary of 22 Studies Using Cytomegalovirus-Specific T Cells as Prophylaxis

Study	Publication Type	Study Design	Population	N	VST Source	Infection Rate, %
Perruccio et al, 2005 [14]	Full paper	Phase 1/2a RCT	HSCT	27	Stem cell donor, autologous	41
Mackinnon et al, 2008a [15]	Full paper	Case series	HSCT	2	Stem cell donor	0
Micklethwaite et al, 2007 [16]	Full paper	Phase 1	HSCT	9	Stem cell donor	22
Micklethwaite et al, 2008 [17]	Full paper	Case series	HSCT	12	Stem cell donor	33
Peggs et al, 2009a [18]	Full paper	Phase 2	HSCT	30	Stem cell donor	30
Peggs et al, 2011a [19]	Full paper	Phase 1/2	HSCT	7	Stem cell donor	14
Hanley et al, 2012 [20]	Abstract	Phase 1	CBT	7	Stem cell donor	14
Blyth et al, 2013 [21]	Full paper	Phase 2	HSCT	50	Stem cell donor	50
Clancy et al, 2013 [22]	Full paper	Phase 1/2	HSCT	7	Stem cell donor	43
Hanley et al, 2013 [23]	Abstract	Case series	HSCT, CBT	42	Stem cell donor	26
Bramanti et al, 2014 [24]	Abstract	Case series	HSCT	8	Stem cell donor, third-party	25
Peggs et al, 2014 [25]	Abstract	Phase 2b RCT	HSCT	20	Stem cell donor	75
Ma et al, 2015 [26]	Full paper	Cohort	HSCT	8	Stem cell donor	100
Naik et al, 2016a [27]	Full paper	Case series	HSCT, CBT	7	Stem cell donor	14
Abraham et al, 2019a [28]	Full paper	Phase 1	CBT	7	Stem cell donor	0
Roex et al, 2020 [29]	Full paper	Phase 1/2	HSCT	24	Stem cell donor	33
Dadwal et al, 2021 [30]	Abstract	Phase 2	HSCT	12	Third-party	0
Gottlieb et al, 2021 [31]	Full paper	Phase 1	HSCT	11	Stem cell donor	91
Rubinstein et al, 2022 [32]	Full paper	Phase 2	HSCT	23	Stem cell donor	17
Gerbitz et al, 2023 [33]	Full paper	Phase 1/2a RCT	HSCT	9	Stem cell donor	44
Jiang et al, 2023 [34]	LTE	Phase 1	HSCT	10	Stem cell donor	50
Kinoshita et al, 2023a [35]	Full paper	Case series	HSCT, CBT	6	Stem cell donor, third-party	0

Abbreviations: CBT, cord blood transplant; HSCT, hematopoietic stem cell transplant; LTE, letter to the editor; RCT, randomized controlled trial; VST, virus-specific T cell.

^aIncluded VSTs used for different indications.

Table 3.

Summary of 38 Studies Using Cytomegalovirus-Specific T Cells as Treatment for Resistant/Refractory Cytomegalovirus Infections

Study	Publication Type	Study Design	Population	N	VST Source	Complete Clearance of DNAemiaa, %	Clearance of Disease, %
HSCT
Einsele et al, 2002 [45]	Full paper	Case series	HSCT	8	Stem cell donor	71b
Cobbold et al, 2005c [36]	Full paper	Cohort	HSCT	2	Stem cell donor	50	NA
Feuchtinger et al, 2010 [46]	Full paper	Case series	HSCT, CBT	18	Stem cell donor, third-party	56b
Bao et al, 2012 [47]	Full paper	Case series	HSCT	7	Stem cell donor	57	NA
Meij et al, 2012 [48]	Full paper	Phase 1/2	HSCT	6	Stem cell donor	100	NA
Koehne et al, 2012 [49]	Abstract	Phase 1	HSCT	13	Stem cell donor	92	NA
Uhlin et al, 2012 [50]	Full paper	Case series	HSCT, CBT	5	Stem cell donor, third-party	80	NA
Leen et al, 2013 [12]	Full paper	Phase 1	HSCT	23	Third-party	39b
Koehne et al, 2015 [51]	Full paper	Phase 1	HSCT	16	Stem cell donor	92	50
Omer et al, 2015 [52]	Abstract	Phase 2	HSCT	10	Third-party	40b
Prockop et al, 2015 [53]	Abstract	Case series	HSCT SOT	11 1	Stem cell donor, third-party	64b
Creidy et al, 2016 [54]	LTE	Case series	HSCT	10	Stem cell donor	30b
Oran et al, 2016 [55]	Abstract	Phase 2	HSCT	11	Third-party	55	NA
Wang et al, 2016 [56]	Abstract	Case series	HSCT	38	Stem cell donor	74	NA
Neuenhahn et al, 2017 [57]	Full paper	Phase 1/2a	HSCT	16	Stem cell donor, third-party	56b
Pei et al, 2017 [58]	Full paper	Cohort	HSCT	32	Stem cell donor	84b
Withers et al, 2017 [59]	Full paper	Phase 1	HSCT	28	Third-party	76	100
Kállay et al, 2018 [60]	Full paper	Case series	HSCT	6	Third-party	67b
Jedema et al, 2019 [61]	Abstract	Phase 1/2	HSCT	14	Stem cell donor	64b
Lozano et al, 2019 [62]	Abstract	Case series	HSCT	6	Stem cell donor	100	NA
Fabrizio et al, 2021 [63]	Full paper	Case series	HSCT, CBT	104	Stem cell donor, third-party	58d
Flower et al, 2020 [64]	Abstract	Case series	HSCT	5	Stem cell donor	100	NA
Keller et al, 2020 [65]	Abstract	Phase 1/2	HSCT	5	Third-party	80	NA
Flower et al, 2023 [66]	Abstract	Phase 1	HSCT SOT	6 1	Stem cell donor, third-party	Not evaluablee
Barba et al, 2022 [67]	Abstract	Phase 1b/2	HSCT	20	Stem cell donor, third-party	70b
Cao et al, 2021 [68]	Abstract	Phase 1	HSCT	7	Stem cell donor	100b
Pei et al, 2022c [41]	Full paper	Case series	HSCT	137	Stem cell donor	87b
Pei et al, 2022 [69]	Full paper	Case series	HSCT	93	Stem cell donor, third-party	88b (stem cell donor) 84b (third-party)
Ruan et al, 2022 [70]	Full paper	Case series	HSCT, CBT	29	Stem cell donor, third-party	90b
Wang et al, 2022 [71]	LTE	Case series	HSCT	10	Stem cell donor	100	100
Heinz et al, 2023 [72]	Full paper	Case series	HSCT	8	Stem cell donor, third-party	100	NA
Jiang et al, 2024 [73]	Full paper	Case series	HSCT	53	Stem cell donor, third-party	85 (stem cell donor) 86 (third-party)	40 (stem cell donor) 50 (third-party)
Pfeiffer et al, 2023 [13]	Full paper	Phase 2	HSCT	24	Third-party	46	33
Prockop et al, 2023 [74]	Full paper	Phase 1/2	HSCT, CBT	59	Third-party	36	30
Keller et al, 2024 [75]	Full paper	Phase 2	HSCT	19	Third-party	54b
Leroyer et al, 2024 [76]	LTE	Case series	HSCT	8	Stem cell donor, third-party	67	40
SOT
Smith et al, 2019 [77]	Full paper	Phase 1	SOT	13	Autologous	15b
Khoury et al, 2024 [78]	Full paper	Phase 2	SOT	26	Third-party	54b

Abbreviations: CBT, cord blood transplant; HSCT, hematopoietic stem cell transplant; LTE, letter to the editor; NA, not applicable; SOT, solid organ transplant; VST, virus-specific T cell.

^aReported specifically for subgroup, if available.

^bReported as response for both infection and disease.

^cIncluded VSTs used for different indications.

^dIncluded both complete and partial clearance.

^eNot evaluable due to data from multiple subgroups reported in aggregate manner.

In total, 48 studies used CMV-only VSTs while the remaining studies used multivalent VSTs. The majority of VSTs were derived from stem cell donors (53/67 studies [79%]), followed by third-party sources in 28 of the studies (42%) and an autologous source in 2 studies (3%). Among third-party VSTs, HLA matching between recipient and VSTs were reported in 16 of 28 studies, and they were matched at 1–6 alleles. Related third-party donors were reported to be used in 3 of 28 studies [66, 72, 73], and the other 25 studies did not identify if third-party donors were related or not. Third-party cells were cryopreserved in 16 of 28 studies and freshly prepared in 11 of 28 studies with limited information in the remaining study. More than half of the studies (34/67 studies) used CMV peptide stimulation to generate VSTs. Direct selection processes using HLA multimers or cytokine-capture techniques to isolate CMV-VSTs were used in 16 of the studies, and 8 studies transduced antigen-presenting cells with a vector expressing CMV peptide to stimulate T cells in vitro. Four studies used both peptide stimulation and vector-induced expression while 5 studies did not report methods of VST generation.

Definitions of CMV infection varied across the studies (Supplementary Table 1). CMV disease was commonly defined as detection of CMV on histopathology, with evidence of tissue invasive disease [13, 26, 45, 73, 74, 77, 78]. Two studies relied on CMV-positive cultures to define CMV disease and 2 studies used detection of nucleic acid [13, 26, 70, 78].

Persistence and functionality of the CMV-VSTs after infusion were evaluated in 31 and 36 studies, respectively (Supplementary Appendix 2). CMV-VSTs were detected up to 24 months after transfusion and the majority of patients had CMV-specific responses as measured by interferon-γ.

CMV Prophylaxis

Patient Characteristics

There were 22 studies that evaluated the use of CMV-VSTs for prophylaxis in HSCT patients, comprising 338 patients (Table 1). No prophylaxis studies included SOT patients. Serostatus was reported in 189 patients from 12 studies. Two studies included only D⁺/R⁺ patients and 1 study included only D⁻/R⁺ patients. The remaining studies included 8%–60% D⁺/R⁻ patients, 33%–92% D⁺/R⁺ patients, and 18%–21% D⁻/R⁺ patients (Supplementary Figure 1A). In 3 studies, 9%–25% of patients were D⁻/R⁻ and received CMV-VSTs as part of multivalent infusions [29, 31, 34].

CMV-VST Source, Infusion, and Antiviral Use

The CMV-VSTs were predominantly derived from stem cell donors in 21 studies, autologous source in 1 study, and third-party sources in 3 studies (Table 1). CMV-VSTs were infused at a median of 45 days (IQR, 36–60 days) following transplant with 4 studies not reporting the timing. In all but 1 study, patients received 1 standard infusion whereas up to 7 infusions were administered in a single study [30].

Outcomes

CMV infection, as defined by respective study authors (Supplementary Table 1), was reported in all 22 studies with a median infection rate of 28% (IQR, 14%–44%) (Table 1). Patients in 4 studies did not experience CMV infection. Figure 2 demonstrates the distribution of infection relative to CMV-VST infusion, with the majority of the infection occurring post infusion. Four studies had patients who also developed CMV disease (range, 12%–27%) affecting the lungs and colon [14, 21, 29, 31].

Figure 2.

Timing of CMV infection relative to CMV-specific T cell infusion used as prophylaxis in 16 studies that had evaluable data.

Mortality due to CMV infection was reported in 19 studies with most of the studies reporting no deaths, and 2 studies reporting a total of 3 deaths from CMV disease [14, 21]. Incidence of worsening or new GVHD was reported in 20 studies with a range of 0–60% following CMV-VST infusion. No cases of cytokine release syndrome (CRS) were reported.

Treatment of Nonrefractory CMV Infection

Patient Characteristics

All of the 15 studies that studied the use of CMV-VSTs for nonrefractory infection were in HSCT patients, constituting 281 patients (Table 2). Donor and recipient serostatus were reported in 7 studies with the majority being D⁺/R⁺ donor–recipient pairs (Supplementary Figure 1B).

Table 2.

Summary of 15 Studies Using Cytomegalovirus-Specific T Cells as Treatment of Nonrefractory Cytomegalovirus Infections

Study	Publication Type	Study Design	Population	N	VST Source	Complete Clearance of DNAemia, %	Clearance of Disease, %
Cobbold et al, 2005a [36]	Full paper	Cohort	HSCT	7	Stem cell donor	100	NA
Peggs et al, 2004 [37]	Full paper	Case series	HSCT	16	Stem cell donor	100	NA
Mackinnon et al, 2008a [15]	Full paper	Case series	HSCT	21	Stem cell donor	Not evaluableb	NA
Peggs et al, 2009a [18]	Full paper	Phase 2	HSCT	20	Stem cell donor	100	NA
Peggs et al, 2011a [19]	Full paper	Phase 1/2	HSCT	11	Stem cell donor	100	NA
Gerdemann et al, 2013 [38]	Full paper	Phase 1/2	HSCT	5	Stem cell donor	80	0
Naik et al, 2016a [27]	Full paper	Case series	HSCT, CBT	5	Stem cell donor	40	NA
Abraham et al, 2019a [28]	Full paper	Phase 1	CBT	4	Stem cell donor	100	NA
Zhao et al, 2020 [39]	Full paper	Cohort	HSCT	35	Stem cell donor	100	NA
Jiang et al, 2022 [40]	Full paper	Phase 1	HSCT	27	Third-party	93	NA
Pei et al, 2022a [41]	Full paper	Case series	HSCT	53	Stem cell donor	96	Not evaluableb
Petitpain et al, 2022 [42]	Abstract	Case series	HSCT	8	Stem cell donor, third-party	38	NA
Galletta et al, 2023 [43]	Full paper	Case series	HSCT	59	Stem cell donor, third-party	56c
Kinoshita et al, 2023a [35]	Full paper	Case series	HSCT, CBT	4	Stem cell donor, third-party	100	100
Ma et al, 2024 [44]	Full paper	Phase 1	HSCT	6	Stem cell donor	100c

Abbreviations: CBT, cord blood transplant; HSCT, hematopoietic stem cell transplant; NA, not applicable; VST, virus-specific T cell.

^aIncluded VSTs used for different indications.

^bNot evaluable due to data from multiple subgroups reported in aggregate manner.

^cReported as response for both infection and disease.

CMV Infection

The time to CMV infection was reported in 4 studies, ranging between 29 and 41 days post-HSCT [39–41, 44]. The majority of the patients had CMV DNAemia with 3 studies reporting patients with CMV disease involving lungs and colon [35, 38, 44]. Only 3 studies reported on the level of DNAemia before VST infusion, ranging between 1.4 × 10³ and 3.1 × 10³ copies/mL, and peak DNAemia ranging between 5.9 × 10³ and 9.5 × 10³ copies/mL [39, 41, 44].

CMV-VST Source, Infusion, and Antiviral Use

The CMV-VSTs were stem cell donor–derived in 14 studies and third-party CMV-VSTs were used in 4 studies. The timing of CMV-VST infusion was reported in 10 studies with a median of 46 days (IQR, 36–52 days) posttransplant. Patients received between 1 and 4 CMV-VST infusions. CMV-VSTs were used as first-line therapy without concomitant antiviral therapy in 3 studies [15, 37, 44], and 14%–100% of patients in 7 studies were already on antivirals at the time of CMV-VST infusion. Seven studies reported the addition of antiviral after CMV-VST infusion, ranging between 29% and 62% of patients. Eight studies reported on the antiviral regimen used, with ganciclovir/valganciclovir in 33%–100% of patients in 7 studies, foscarnet in 17%–100% of patients in 5 studies, cidofovir in 25%–50% of patients in 3 studies, and CMV immunoglobulins in 20%–25% of patients in 3 studies.

Outcomes

Complete responses and partial responses were variably defined (Supplementary Table 2). Fourteen studies reported on complete responses with evaluable data, with the median being 98% (IQR, 70%–100%). Two studies reported on partial responses, with a range of 7%–40%. Time to CMV DNAemia clearance was reported in 4 studies, ranging between 4 and 21 days post–VST infusion [39–41, 44]. Seven studies reported recurrent CMV infections occurring in 0–25% of patients. There were no CMV-related deaths reported. Incidence of worsening or new GVHD was reported in 11 studies, ranging between 0 and 36%. One episode of CRS was reported among 10 studies explicitly reporting CRS [44].

Treatment of R/R CMV Infections in HSCT

Patient Characteristics

There were 36 studies that evaluated the use of CMV-VSTs in R/R CMV infections in HSCT recipients, comprising 867 patients (Table 3). CMV serostatus was reported in 9 studies (Supplementary Figure 1C).

CMV Infection and Disease

CMV DNAemia levels were reported in 16 studies; peak DNAemia ranged from 6.2 × 10³ IU/mL to 1.2 × 10⁵ copies/mL. CMV disease was reported in 14 studies with pneumonitis reported in 11 studies, central nervous system involvement in 5 studies, retinitis in 5 studies, and colitis in 10 studies. Time to CMV infection was reported in 12 studies and occurred at a median of 34 days (IQR, 29–36 days) post-HSCT.

CMV-VST Source, Infusion, and Antiviral Use

CMV-VSTs were derived from the HSCT donor in 27 studies and third-party donors in 21 studies. Additional viral targets were used in 7 multivalent studies. The number of VST infusions ranged from 1 to 15. A median time to VST infusion was reported as 81 days (IQR, 69–110 days) post-HSCT by 15 studies. Antiviral medication was used in 89%–100% of cases. Ganciclovir or valganciclovir use was reported in 43%–100% of patients in 12 studies, 11 studies reported 6%–88% of patients receiving foscarnet, 8 studies reported 13%–65% of patients receiving cidofovir, and 4 studies reported 71%–88% of patients receiving CMV immunoglobulins. Acyclovir and maribavir were used in 1 study each.

Outcomes

The median complete response rate was 70% (IQR, 56%–88%) (Figure 3). Partial response rates were reported in 10 studies at a median of 29% (IQR, 19%–39%). CMV-related mortality was reported in 27 studies with a total of 54 deaths. Worsening or new GVHD was reported in 23 studies, ranging between 0 and 20% post–VST infusion. Incidence of CRS was reported in 24 studies with the majority reporting none, but 1 study reported 1 patient experiencing grade 1/2 CRS [68]. Three episodes of secondary graft rejection were reported in a single study; 1 episode was related to third-party VST infusion with marked expansion of VST-donor T cells and loss of chimerism [75, 79].

Figure 3.

Resistant/refractory CMV infection response rate after CMV-specific T cell infusion. ^aSubgroup that used third-party VSTs.

Treatment of R/R CMV Infections in SOT

Patient Characteristics

Four studies included the use of CMV-VSTs in SOT recipients with R/R CMV infections (41 patients) [53, 66, 77, 78]. Two of the SOT studies reported on the organ type with 14 kidney recipients, 6 liver recipients, 4 heart recipients, 14 lung recipients, and 1 small bowel/multivisceral recipient [77, 78]. Donor/recipient serostatus was only reported in 1 study, with 62% D⁺/R⁻, 23% D⁺/R⁺, and 15% D⁻/R⁻.

CMV Infection and Disease

There were limited data on the level of DNAemia and timing of CMV infection relative to transplant. Anatomical disease site was reported in 2 studies involving central nervous system, retinitis, pneumonitis, enteritis, colitis, and hepatitis [53, 77].

CMV-VST Source, Infusion, and Antiviral Use

CMV-VSTs were derived from SOT donors in 2 studies, from third parties in 3 studies, and autologously in 1 study. One study reported on the timing of CMV-VST infusion, administered at a median of 724 days (IQR, 266–979 days) posttransplant. Patients received between 2 and 6 infusions. One study provided details on antiviral use, with 100% of patients on valganciclovir/ganciclovir, 62% on foscarnet, 8% on cidofovir, and 31% receiving CMV immunoglobulins, with 15% of patients continuing and 15% requiring additional antivirals post-VSTs.

Outcomes

Response rates were evaluable in 3 studies, ranging between 15% and 64% complete response and between 9% and 38% partial response. One study indicated that 8% of patients had recurrent CMV infection. Data on DNAemia clearance were limited. There was 1 CMV-related death reported [77]. There were no cases of CRS in the 3 studies that reported on this. One study that administered multivalent VSTs reported 3 instances of acute organ rejection (2 kidneys and 1 liver) within 1 month following infusion [78].

DISCUSSION

This scoping review represents the most comprehensive synthesis of available literature on the clinical experience of CMV-VST use across different indications in adult and pediatric HSCT and SOT populations. Patient characteristics, antiviral regimens, VST product, and clinical response were heterogeneously reported in the included 67 articles.

Overall, most data are derived from CMV-VST use in HSCT patients reported in retrospective studies and phase 1/2 trials. Important definitions of CMV infection and disease were variable across studies (Supplementary Tables 1 and 2). In the context of non-R/R CMV infection, most patients were CMV D⁺/R⁺, and responded well to CMV-VST with a complete response in excess of 90%. Details on CMV levels before and postinfusion, antiviral discontinuation, and need for additional antivirals were infrequently reported, rendering the evaluation of the independent effect size of CMV-VSTs a challenge. In the R/R CMV infection setting, there was an increased proportion of D⁻/R⁺ patients reported, consistent with this group being the highest risk group for CMV infection in HSCT. The median complete response was 70% (IQR, 56%–88%), with the majority of patients receiving concurrent antivirals, confounding the effectiveness of VSTs in the absence of control groups. Responses were similar between original graft donor VSTs and third-party VSTs in R/R CMV infections when 2 or more HLA alleles were matched [63, 69, 72, 73]. Notably, the majority of studies were conducted in the era before letermovir and maribavir were approved for CMV prophylaxis post-HSCT and treatment of R/R CMV infection and disease, respectively. The role of CMV-VST in the context of contemporary antiviral therapies remains to be defined. Ideally, prospective head-to-head clinical trials would characterize the true efficacy and safety of each therapeutic approach. Furthermore, scarce long-term outcome data are available to assess the impact of VST therapy on subsequent viral reactivations and overall outcomes.

VSTs were generally well-tolerated with only 2 studies reporting 2 patients experiencing maximum grade 2 CRS. Worsening or new GVHD was reported in up to 60% of patients following VST infusion with the highest rates in the prophylaxis group, when patients are in the highest risk period of developing acute GVHD. The contributory effects of VST to incidence of CRS and GVHD remain to be determined. Episodes of allograft rejection associated with VST were rare, except for 1 reported case of VST use following HSCT where there was loss of donor chimerism and expansion of VST T cells [79].

SOT recipients represent a lesser fraction of the studies evaluating CMV-VSTs for refractory CMV infections, with complete response rates varying between 15% and 64%. This may be due to challenges in accessing donor T cells for manufacturing compared to the HSCT population, time required for VST generation, and limited expansion of the infused CMV-VSTs in the recipient in the setting of ongoing immunosuppression and lack of preinfusion myeloablative conditioning. Several phase 1/2 trials that include SOT patients are ongoing and will provide further insights into the feasibility of using VSTs, especially third-party, in this patient population [80].

In principle, CMV-VSTs are an appealing alternative to traditional CMV pharmacotherapy. However, heterogenous reporting of key variables across the studies makes the interpretation and clinical translation of CMV-VSTs challenging. Standardization of CMV DNA quantification and CMV-related definitions are needed to improve the quality of CMV-VST. Universal adoption of the updated 2024 CMV definitions across studies will enable proper comparison [81]. Furthermore, automated and scalable production of CMV-VSTs is necessary for widespread uptake to be feasible. Commercial banked VSTs could alleviate this issue; however, the largest and most advanced studies that used this approach were discontinued due to futility [82]. Additional knowledge gaps that must be addressed include the impact of preceding and concurrent antiviral treatment on VST efficacy, as well as the optimal number of infusions, dosing, and timing. A global registry that standardizes key data elements to be tracked over time may help unify this process and enable meta-analysis of the clinical efficacy of CMV-VSTs. With >30 years since the emergence of the VST proof-of-concept study, the lack of high-quality interventional trials demonstrating CMV-VST efficacy warrants a reconsideration of the biological plausibility and feasibility of this approach.

There are several limitations to this scoping review. First, case reports and series with <5 patients were excluded, which may have led to underrepresentation of patient populations, specifically SOT patients. In addition, it may be possible that the results of select patients are reported in >1 study as part of a bigger cohort of patients.

CONCLUSIONS

CMV-VSTs are increasingly being used in HSCT patients compared to SOT patients. The scalability of these products and extrapolation of their use across different patient populations remain to be determined, specifically in the current antiviral era. Consensus-based recommendations may be considered by the transplant community to establish the standard reporting of relevant characteristics and outcomes of patients undergoing interventional trials with VSTs for empirical evaluation of their clinical efficacy.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. J. S. T. F. and R. C. W.: Data collection, analysis, and interpretation, drafting. D. K. B.: Data collection. D. H.: Search strategy. A. A., J.-S. D., L. S., and R. K. A.: Critical revisions. S. B.: Conception, drafting, and critical revisions. All authors approved the final version of the manuscript.

Financial support. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.