The CCP-COVID-19-HSCT-2023 trial (ClinicalTrials.gov NCT05904067) was a prospective, single-center, randomized controlled trial conducted at the Hematopoietic Stem Cell Transplantation Center, Institute of Hematology & Blood Diseases Hospital, between June 2023 and February 2024. Participants were randomly assigned in a 1:1 ratio to the CCP group or standard treatment group using a centralized randomization system and were followed for 12 months. A detailed description of the study design, intervention schedule, and safety monitoring procedures can be found in the full protocol (Supplementary Appendix 1).

The randomization sequence was computer-generated by an independent statistician using a simple randomization algorithm and was not accessible to the clinical investigators. Allocation concealment was maintained through sealed, sequentially numbered opaque envelopes prepared by the data manager, which were opened only after each participant had provided written informed consent and eligibility was confirmed.

This process ensured that investigators enrolling patients could not predict group assignment prior to randomization. The study was conducted in an open-label design. To minimize potential bias, laboratory personnel conducting SARS-CoV-2 PCR testing and antibody-titer assays were blinded to group allocation. In addition, statistical analyses were performed on a de-identified dataset with coded treatment groups until data lock. Investigators and participants were aware of treatment assignment only for clinical management and safety monitoring purposes.

The trial was conducted in an open-label design, as blinding of plasma infusion was not feasible. To minimize potential bias, laboratory personnel conducting SARS-CoV-2 PCR testing and antibody-titer assays were blinded to group allocation. In addition, investigators and participants were aware of treatment assignment only for clinical management and safety monitoring purposes.

The trial protocol was approved by the Institutional Review Board (IIT2023001-EC-1). Written informed consent was obtained from all participants before enrollment. The study adhered to the Declaration of Helsinki and good clinical practice guidelines.

Eligible participants were adults (≥16 years) undergoing HSCT. Exclusion criteria included a history of COVID-19, SARS-CoV-2 positivity within 13 days post-transplantation, known HIV, active HBV or HCV, psychiatric disorders, severe plasma protein allergies, or IgA deficiency. Detailed criteria are in the study protocol (Supplementary Appendix 1).

The CCP group received 200 mL of COVID-19 CCP (neutralizing antibody titer ≥1:160) on days +14, +28, +2 months, and +3 months post-transplantation, alongside standard treatment (see Supplementary Table 1 for detailed references and pathway mapping). CCP was sourced from COVID-19-recovered donors (negative nucleic acid tests) within two weeks, as fresh or frozen plasma.

Anti-SARS-CoV-2 IgG in donor plasma was measured using the WANTAI SARS-CoV-2 IgG ELISA kit (Beijing Wantai Biological Pharmacy Enterprise Co., Ltd., China), following the manufacturer’s instructions. This assay provides semiquantitative signal-to-cutoff (S/CO) ratios and is intended for qualitative determination of antibody positivity. According to the standards used by national blood centers in China, plasma with a neutralizing antibody titer ≥1:160 was defined as high-titer CCP. The assay is not calibrated to the WHO International Standard, and conversion to binding antibody units per milliliter (BAU/mL) was therefore not feasible. All antibody data in this study are thus reported as semiquantitative results based on the manufacturer’s criteria (details in Supplementary Material).

Both study arms received identical post-transplant supportive care, including antimicrobial prophylaxis, GVHD prevention, and hepatoprotective therapy. UDCA was routinely administered to all participants (250 mg three times daily) according to institutional protocol to prevent hepatic injury and cholestasis. UDCA was not considered part of the study intervention, and its use and dosage were consistent across both groups. Previous evidence from our institution confirmed that UDCA has no antiviral or anti–SARS-CoV-2 effect (7).

Adverse events were graded using CTCAE v5.0 and EBMT guidelines, with toxicity relationships categorized as certain, probable, possible, or unlikely per WHO criteria.

Data were collected via electronic medical records. Participants underwent regular SARS-CoV-2 nucleic acid and antibody testing. Severe cases prompted chest CT, liver and kidney function tests, and arterial blood gas analysis for oxygen saturation <94%.

The primary outcome was the incidence of laboratory-confirmed COVID-19 infection from the first CCP infusion to 28 days after the last infusion, corresponding approximately to day +120 after transplantation. This window covers the entire prophylactic period of CCP administration (days +14, +28, +60, and +90). Secondary outcomes included severe infections, 30-day survival post-infection, one-year OS, and adverse reactions to CCP (definitions in Supplementary Material). Outcomes align with the 120-day post-transplantation timeframe for data presentation.

A pre-specified analysis plan included ITT analyses to ensure comprehensive evaluation. The sample size was determined based on the expected difference in COVID-19 infection rates between the two study arms. The infection rate within one year after HSCT was estimated at 80% in the control group and 50% in the CCP group, based on preliminary institutional data and published reports. Using a two-sided significance level (α = 0.05) and a power of 80% (β = 0.20), and applying a log-rank test for group comparison, the required total sample size was calculated to be 72 participants (36 per group). A 10% attrition rate was incorporated to account for potential losses to follow-up. The calculation was performed using PASS version 15.0 (NCSS, Kaysville, UT, USA). The sample size and methods are detailed in the Supplementary Appendix 1. Continuous variables were compared using the Mann–Whitney U-test, and categorical variables using Chi-squared test or Fisher’s exact test. Kaplan–Meier plots and log-rank tests were used for time-to-event analyses. To evaluate the 120-day cumulative infection rate, we used Royston-Parmar flexible parametric survival models, adjusting for confounders (age, sex, and HCT-CI scores). All results are reported with corresponding 95% confidence intervals (CIs) and p values. Statistical significance was defined as a two-sided p < 0.05 after correction, unless otherwise specified. Statistical analyses were performed using R software, version 4.1.3, and SPSS 20.

Between June 2023 and February 2024, 313 patients undergoing HSCT were screened, with 72 randomized 1:1 to the standard treatment group (n=36) or the CCP group (n=36) (Figure 1). Three patients from the CCP group withdrew before receiving any infusion, resulting in an overall withdrawal rate of 4.2%. Among the 33 CCP recipients, 26 (26/33, 78.8%) completed all four cycles. Infusion was discontinued in 4 (4/33, 12.1%) patients due to severe aGVHD, pulmonary infection, or hematological relapse, and three (3/33, 9.1%) missed scheduled infusions due to non-adherence. In this study, 72 patients were analyzed under intention-to-treat (ITT) principles. The baseline characteristics were shown in Table 1.

19 developed confirmed Omicron variant COVID-19 within 120 days post-transplant in this study. The median time from transplantation to infection occurrence was 95 (IQR, 81-106) days in the CCP group and 74 (IQR, 48-110) days in the standard treatment group (p = 0.272). Kaplan-Meier estimates of the 120-day cumulative incidence of infection showed no statistically significant difference between the CCP and standard groups (32.1% [95% CI, 14.5-46.1%] vs. 22.9% [95% CI, 7.6-35.6%]; p=0.459) (Figure 2). Despite numerically lower infection rates in the CCP group at early timepoints, no statistically significant protection was observed through 120 days (Figure 2). Besides, adjusted estimated risks at 120 days were 26.4% (95% CI, 12-40.8%) for the CCP group and 18.1% (95% CI, 5.5-30.7%) for the standard group (Difference, 8.3% [95% CI, -10.8%-27.4%]; p = 0.400; Supplementary Figure 1A). The 120-day cumulative incidence of infection was analyzed according to the number of plasma infusion cycles. There was no significant difference among patients who received four cycles of infusion, fewer than four cycles, or standard treatment (p = 0.365; Supplementary Figure 2).

Two patients developed severe COVID-19 infection, with one case in each group [2.8% (95%CI, 0-8%)] for both CCP and control groups. In the CCP group, the event occurred at 72 days post-transplantation after 3 infusions, whereas the control group case occurred at 116 days. The detailed information of patients in the CCP group was shown in Supplementary Table 2. The Kaplan-Meier estimated cumulative incidence of severe COVID-19 did not differ significantly between group (Figure 3A; p = 0.992). Neither of the two patients died within 30 days of infection.

During the follow-up period, a total of 12 (12/72, 16.7%) patients died, with 9 deaths (9/36, 25.0%) occurring in the CCP group and 3 deaths (3/36, 8.3%) in the standard group. The detailed causes of death were shown in Table 2. The one-year OS rate was significantly lower in the CCP group compared to the standard group (73.1% [95% CI, 59.4-90.0%] vs. 91.7% [83.1-100%]; p=0.044, Figure 3B).

To assess whether the number of plasma infusion cycles affected survival, we performed a stratified analysis. In the CCP group, patients who received four cycles had a 1-year OS of 79.2% (95% CI, 64.4-97.4%), comparable to the standard treatment group [91.7% (95% CI, 83.1-100%); p = 0.182]. In contrast, patients who received fewer than four cycles had a significantly lower 1-year OS of 57.1% (95% CI, 32.6-100%; p = 0.013) (Supplementary Figure 2). Among the 10 patients who did not complete the full course of plasma infusion, 4 died, including 2 from disease relapse, 1 from severe pneumonia unrelated to COVID-19, and 1 from severe intestinal infection and GVHD.

Among the CCP group, 5 patients (5/36, 13.9%) suffered hematological relapsed, while in the standard group, only 2 patients experienced extramedullary relapse (2/36, 5.6%) within one year post-transplantation. Of the 5 patients in the CCP group, 3 completed 4 cycles of plasma infusion, with relapse occurring at 135, 176, and 262 days post-transplantation, respectively. One received 2 cycles of plasma infusions and relapsed at 57 days post-transplantation. The remaining one did not receive any plasma infusions and relapsed at 122 days post-transplantation.

Post-transplant immune reconstitution analysis was performed. Overall, there were no significant differences in the proportions of T cells, B cells, and NK cells in the peripheral blood (Supplementary Figure 3). However, at 90 days post-transplantation, the proportion of CD4+ T cells among lymphocytes and the CD4/CD8 ratio in the CCP group were significantly lower than those in the standard group (Figure 4; p = 0.012 and p = 0.008, respectively).

Among the 33 patients in the CCP group who received at least one infusion, adverse events were observed in 3 cases (3/33, 9.1%). The most common adverse event was fever, reported in 2 patients (2/33, 6.1%), followed by rash in 1 patient (1/33, 3.0%). All events resolved within 24 hours with supportive care, and no serious adverse events were reported (Table 3; Supplementary Table 3).

All plasma units used for infusion met the national criterion for high-titer convalescent plasma (≥1:160 by ELISA). Antibody absorbance (optical density, OD) and neutralizing titers were measured at predefined intervals post-HSCT. Antibody testing was semiquantitative and not directly comparable to WHO-standardized neutralizing titers. Median absorbance levels were consistently higher in the CCP group at Days 28, 60, and 90, but without statistical significance (Supplementary Figure 4A). OD levels were also lower in infected compared to non-infected patients up to Day 21, though the differences were not significant (Supplementary Figure 4B, C).

Neutralizing antibody seropositivity (titer ≥1:160) remained low across groups. Further details were available in Supplementary Table 4. The cumulative infection rate at 120 days post-transplant was significantly lower in antibody-positive patients than in antibody-negative patients (14.0% [95% CI, 1.8-24.7%] vs. 41.2% [95%CI, 22.1-55.6%]; p = 0.016). In addition, antibody titer levels had no significant impact on 1-year OS (83.4% [95% CI, 72.1-96.5] for antibody-positive vs. 82.1% [95% CI, 70.1-96.2] for antibody-negative; p = 0.896; Supplementary Figure 2).

To evaluate the efficacy of CCP across different subgroups, we performed analyses stratified by age, gender, type of transplant, and HCT-CI scores for the 120-day cumulative infection rate (Supplementary Figure 5). No significant interactions were observed in any subgroup. However, a non-significant trend toward better efficacy was noted in females compared to males (p = 0.089), warranting further investigation.