ABSTRACT
Allogeneic hematopoietic stem cell transplantation (allo- HSCT) has traditionally involves administering fresh peripheral blood or bone marrow stem cells. At onset of the COVID-19 pandemic in March 2020, the National Marrow Donor Program (NDMP) mandated cryopreservation of all unrelated peripheral blood stem cell (PBSC) products to prevent interruptions in transplant plans by donor COVID-19 infection after recipient's start of conditioning chemotherapy. Since the lifting of this mandate, many centers have continued to cryopreserve grafts prior to initiation of conditioning, but the longer-term clinical outcomes of this practice including chronic graft versus host disease (cGVHD) rates of patients receiving cryopreserved stem cells have not been previously well described. Prior work has raised concern for a deleterious effect of cryopreservation on overall survival and non-relapse mortality (PMID: 33865804). However, heterogeneity in the patient population and reason for cryopreservation suggest that further study is needed to assess these outcomes. Here we report our single-institution experience of clinical outcomes using cryopreserved versus fresh URD PBSCs for allo-HSCT. We examined long-term outcomes in 387 patients who received unrelated donor (URD) PBSCs (136 cryopreserved, 251 fresh) between January 1, 2019 and July 31, 2021. The cohorts had similar baseline characteristics including donor/recipient age/sex, disease, conditioning regimen/intensity, and GVHD prophylaxis regimens. Two-year OS, PFS, relapse, NRM, and acute GVHD rates were not different between recipients of fresh versus cryopreserved PBSCs. Strikingly, 2-year incidence of cGVHD (28% vs 52%, p=0.00001) and moderate/severe cGVHD (9% vs 24%, p=0.00016) was substantially lower in recipients of cryopreserved PBSCs compared to fresh, respectively (Figure 1). This difference was only noted in patients receiving a GVHD prophylaxis regimen without post-transplantation cyclophosphamide (PTCY) (no PTCY 2-year cGVHD incidence cryopreserved vs fresh: 29% vs 57%, p=0.000016), moderate/severe cGVHD 16% vs 34%, p=0.0006) (Figure 2). For patients receiving a PTCY-containing GVHD prophylaxis regimen, there was no difference in cGVHD incidence (cGVHD cryopreserved vs fresh: 24% vs 27%, p=0.56, moderate/severe cGVHD 7% vs 9.3%, p=0.3, Figure 3). (Figure Presented) (Figure Presented) (Figure Presented) While survival and relapse rates are not different, cryopreservation is associated with a marked reduction in cGVHD rates in the setting of non-PTCy based GVHD prophylaxis. Larger multicenter or registry analyses are needed to confirm these observations and may prompt a re-assessment of the role of cryopreservation of stem cell products in clinical practice. If confirmed, it will be critical to understand the immunologic consequences of cryopreservation and how they might influence the clinical impact on chronic GVHDCopyright © 2023 American Society for Transplantation and Cellular Therapy
ABSTRACT
Background: Immunocompromised persons are disproportionately affected by severe SARS-CoV-2 infection, but immune compromise is heterogenous, which may impact viral dynamics. We hypothesized that higher degrees of compromised immunity are associated with higher viral shedding and slower viral clearance in the absence of COVID-19 therapeutics. Method(s): Participants enrolled in ACTIV-2/A5401, a platform trial for COVID-19 therapeutics in non-hospitalized adults within 10 days of symptom onset, received either an active treatment or placebo between 8/2020 and 7/2021. Participants were categorized based on the extent of immunosuppression into none, mild, moderate and severe categories at enrollment (day 0). Longitudinal anterior nasal (AN) and plasma SARS-CoV-2 levels were measured with a quantitative PCR assay. Regression models assessed associations between immunocompromise severity and viral levels (VL) at day 0, and longitudinally among those on placebo with quantifiable RNA at day 0. Multivariate analyses adjusted for demographics and symptom duration and vaccination status at day 0. Result(s): Immunocompromised (mild 383, moderate 159, severe 35) and immunocompetent (1956) participants had comparable symptom durations at day 0 (median 6 days) and most were unvaccinated (~95%). AN VL at day 0 was higher in the moderate/severe group compared to the immunocompetent group (adjusted difference in means: 0.47 log10 copies/mL, 95% CI 0.12, 0.83). While AN VL decayed at similar rates among all groups from day 0 to 3, there was a trend towards higher cumulative AN VLs across the 28-day follow-up in the moderate/severe group compared to immunocompetent group (adjusted fold difference in VL AUC 1.63, 95%CI 0.95, 2.77). The mild group showed no differences in day 0 VL or AUC compared to the immunocompetent group. The frequency of detectable plasma SARS-CoV-2 RNA was similar at day 0 across all groups (overall 21%), but there appeared to be a higher proportion of immunocompromised participants with detectable plasma viral RNA at day 7 (moderate/severe 2/23 [9%], mild 5/44 [11%]) compared to the immunocompetent group (8/282, 3%). Conclusion(s): Before emergence of Omicron and widespread vaccination, moderate/severe immunocompromised status was associated with higher nasal viral levels at study enrollment and showed a trend towards higher cumulative AN viral load, and all immunocompromised groups appeared to have more persistent plasma viremia during follow-up.
ABSTRACT
Background: Reliable biomarkers of COVID-19 severity and outcomes are critically needed for clinical and research applications. We evaluated associations between anti-Spike IgG and SARS-COV-2 nucleocapsid antigen (N Ag) in plasma with clinical outcomes in outpatients with COVID-19. Method(s): We used data from 229 non-hospitalized, US-based adults with COVID-19 who enrolled between January and July 2021 into the placebo arm of the ACTIV-2/A5401 platform trial within 10 days of symptom onset. Pretreatment (day 0) plasma was analyzed by the quantitative Simoa SARS-CoV-2 IgG antibody (anti-Spike) assay (lower limit of quantification [LLoQ] 0.77ug/ mL), and the quantitative Simoa SARS-CoV-2 N Protein Advantage (Quanterix) measuring N Ag (LLoQ 3pg/mL). In addition to analyses for < LLoQ vs >=LLoQ anti-Spike and N Ag, we categorized participants into five N Ag groups (< 3 pg/ml;3-< 100 pg/ml;100-< 1,000 pg/ml;1,000-< 2,500 pg/ml;>=2,500 pg/ ml). Associations between SARS-CoV-2 anti-Spike and N Ag levels and clinical outcomes (all-cause hospitalization/death through day 28 and time to symptom improvement or resolution for two consecutive days from day 0 status) were estimated using log-binomial and Cox regression models, respectively. Result(s): At day 0, 40% had anti-Spike levels >=LLoQ and 64% of participants had plasma N Ag levels >=LLoQ. Participants with anti-Spike levels < LLoQ compared to those who had quantifiable anti-Spike at day 0, had an increased risk of hospitalization/death (16% vs 2%, RR [95% confidence interval (CI)]: 7.3 [1.8, 30.1]), and a significantly longer time to symptom improvement (median [Q1, Q3] 14 days [8, >27] vs 9 days [4, 16], hazard ratio [HR]: 0.6 [95%: CI: 0.4, 0.8], p< 0.001). Participants with higher N Ag levels at day 0 had an increased risk of hospitalization or death, ranging from 1% for < 3 pg/ml to 70% for >=2500 pg/ml (Figure). Compared to individuals who had N Ag levels < LLoQ at day 0, those in the highest category of N Ag levels (>=2500 pg/mL) experienced a significantly longer time to symptom improvement (median [Q1, Q3]: 25 days [13, >27] vs 10 days [5, 20];HR: 0.4 [95% CI: 0.2, 0.7];p=0.04). Conclusion(s): At study entry, the absence of Spike antibodies and higher levels of plasma SARS-CoV-2 N Ag predicted hospitalizations and death in untreated outpatients with COVID-19. These parameters may serve as informative biomarkers for risk stratification in the evaluation of outpatients with COVID-19. (Figure Presented).
ABSTRACT
Background: Whether early antiviral therapy reduces the risk of Long COVID is not known. The combination SARS-CoV-2 monoclonal antibodies amubarvimab+romlusevimab (A+R) were highly effective in reducing 28-day all-cause hospitalization/death among high-risk adults with mild-to-moderate COVID-19 in the randomized, placebo-controlled ACTIV-2/A5401 trial. We assessed the impact of A+R on late outcomes including Long COVID in ACTIV-2. Method(s): A long-term (LT) symptom diary and 2 health-related quality of life questionnaires (EQ-5D-5L and SF-36v2) were completed at week 36. The primary analysis compared the proportion of participants with the composite outcome of self-reported Long COVID (having any COVID-19 symptoms present on a global assessment question in LT diary) at week 36, or hospitalization or death by week 36 between A+R and placebo using regression models with inverse probability weighting to account for incomplete outcome data;supplemental analysis compared the proportion with Long COVID among those alive. Other analyses were restricted to observed data only. Result(s): 807 were randomized and received A+R (n=405) or placebo (n=402) from Jan-July 2021. At entry, median age was 49 years, 51% were female, >99% cis-gender, 17% Black/African American, 50% Hispanic/Latino, and 9% previously received COVID vaccination. 70 (17%) on A+R and 93 (23%) on placebo met the primary outcome;113 (14%) had incomplete data for determining the outcome (Figure 1). Accounting for incomplete data, weighted Risk Ratio [wRR]=0.74;95% CI: 0.56, 0.97;p=0.03. The difference was driven by fewer hospitalizations/deaths in the A+R arm (5%) than placebo arm (15%), particularly by day 28. Excluding 12 participants who died by week 36, frequency of Long COVID was similar in the arms, 16% for A+R and 14% for placebo (wRR=1.09;95%CI: 0.75, 1.58;p=0.64). There were no differences in the proportions reporting return to pre-COVID health (global assessment) or individual symptoms, or in number of symptoms reported or distribution of worst symptom severity. RRs favored the A+R arm on several EQ-5D-5L domains, but none met statistical significance. No differences were observed on SF-36v2 assessments. Conclusion(s): While A+R was highly effective in preventing all-cause hospitalizations and deaths in high-risk outpatients with mild-to-moderate COVID-19, there was no meaningful effect of treatment on measures of Long COVID at 36 weeks. Additional interventions are needed for Long COVID prevention. (Figure Presented).
ABSTRACT
Background: Rebound of SARS-CoV-2 RNA and symptoms has been reported in people treated with nirmatrelvir/ritonavir. Since the natural course of viral and symptom trajectories during COVID-19 have not been well described, we evaluated the incidence of viral rebound and symptom relapse in untreated individuals with mild-to-moderate COVID-19. Method(s): This analysis included 563 participants randomized to placebo in the ACTIV-2/A5401 platform trial. Participants recorded the severity (scored as 0-3) of each of 13 targeted symptoms daily from days 0-28, with symptom score being the summed score (0-39). Symptom rebound was defined as >=4 point increase in symptom score between the maximum and the preceding minimum score. Anterior nasal (AN) swabs were collected for SARS-CoV-2 RNA testing on days 0-14 and 28. Viral rebound was defined as a >=0.5 log10 RNA copies/mL increase from the immediately preceding time point to a level >=3.0 log10 RNA copies/mL, with high-level rebound defined as an increase of >=0.5 log10 copies/mL to a level >=5.0 log10 RNA copies/mL. To mirror the timing of a 5-day nirmatrelvir/ritonavir course, a supportive analysis was conducted where participants were only classified as rebounders if their rebounds occurred on or after day 5. Result(s): Symptom rebound was identified in 26% of participants at a median [Q1, Q3] of 6 [4, 9] days after study entry and 11 [9, 14] days after initial symptom onset. Individuals with symptom rebound were more likely to be female, at high risk for progression to severe disease, have shorter time since symptom onset at study entry, and have higher symptom score and higher AN viral levels day 0. Viral rebound was detected in 32%, with high-level rebound in 13% of participants. Participants with viral rebound were older, more likely to be at low risk for progression to severe disease and had higher median AN viral level at day 0. Most symptom and viral rebound were transient with 89% of symptom rebound and 95% of viral rebound events occurring for only a single day before improving. The combination of symptom and high-level viral rebound was observed in 3% of participants. In the supportive analysis of rebound occurring >=5 days after study entry, 22% and 20% of participants met symptom and viral rebound criteria, respectively, but only 1.2% of participants met criteria for both symptom and high-level viral rebound. Conclusion(s): Symptom or viral rebound in the absence of antiviral treatment is common, but the combination of symptom and viral rebound is rare.
ABSTRACT
Background: Amubarvimab and romlusevimab are anti-SARS-CoV-2 monoclonal antibodies (mAbs) that significantly reduced the risk of hospitalizations or death in the ACTIV-2/A5401 trial. SARS-CoV-2 variants (e.g., Delta, Epsilon, Lambda) harbor mutations against romlusevimab. We evaluated viral kinetics and resistance emergence in individuals treated with mono versus dual-active mAbs. Method(s): The study population included 789 non-hospitalized participants at high risk of progression to severe COVID-19 enrolled in the ACTIV-2/ A5401 platform trial (NCT04518410) and received either placebo (n=400) or amubarvimab plus romlusevimab (n=389). Anterior nasal (AN) swabs were collected for SARS-CoV-2 RNA testing on days 0-14, and 28. Spike (S) gene nextgeneration sequencing were performed on samples collected at study entry and the last sample with viral load >=2 log10 SARS-CoV-2 RNA copies per ml. We compared viral load kinetics and resistance emergence with single versus dual-active mAbs by categorizing participants as harboring variants sensitive to amubarvimab alone (Delta, Epsilon, Lambda, Mu) versus those sensitive to both mAbs (Alpha, Beta, Gamma, Others). Result(s): Study participants receiving single and dual-active mAbs had similar demographics, baseline AN viral load, baseline symptom score and duration since symptom onset. The most common SARS-CoV-2 variant in the study population was Delta (26%) followed by Gamma (19%), Alpha (12%), and Epsilon (10%). In those with successful sequencing, 37% (N=111) were infected with a variant sensitive to amubarvimab alone and 63% (N=188) were infected with a variant sensitive to both mAbs. Compared to treatment with a singleactive mAb, treatment with dual-active mAbs led to faster viral load decline at study day 3 (p=0.0001) and day 7 (p=0.003). Treatment-emergent resistance mutations were significantly more likely to be detected after amubarvimab plus romlusevimab treatment than placebo (2.6% vs 0%, P=0.0008). mAb resistance was also more frequently detected in the setting of single-active mAb treatment compared to dual-active mAb treatment (7.2% vs 1.1%, p=0.007). Participants with emerging mAb resistance had significantly higher pretreatment SARS-CoV-2 nasal viral RNA levels. Conclusion(s): Compared to single-active mAb therapy, dual-active mAb therapy led to significantly faster viral load decline and lower risk of emerging mAb resistance. Combination mAb therapy should be prioritized for the next generation of mAb therapeutics.
ABSTRACT
Background. Predictors of SARS-CoV-2 RNA levels and changes over time during early COVID-19 are not well characterized. Methods. ACTIV-2 is a phase II/III randomized, placebo-controlled, platform trial to evaluate investigational agents for treatment of COVID-19 in non-hospitalized adults. Participants enrolled within 10 days of symptom onset. Nasopharyngeal samples were collected for SARS-CoV-2 RNA testing on Days 0, 3, 7, 14 and 28;RNA was quantified with qPCR assay. SARS-CoV-2 seropositivity was defined as detectable IgG to any of nucleocapsid, receptor binding domain, S1 and S2 antigens by Bio-Plex multiplex assay. Censored linear regression and repeated measures Poisson models evaluated predictors of RNA including age, sex, race, ethnicity, risk of severe COVID-19, diabetes, BMI, obesity (BMI > 35 kg/m2) and serostatus. Results. The study enrolled 537 participants from Aug 2020 to July 2021 at US sites. Median age was 48 years;49% were female sex, >99% cis-gender, 83% white, 29% Hispanic/Latino, and 21% had BMI > 35 kg/m2. At Day 0, median symptom duration was 6 days, 50% were seropositive (2 were vaccinated) and 17% had RNA below the lower limit of quantification (LLoQ). Higher Day 0 RNA was associated with shorter symptom duration (Spearman correlation = -0.40, p< 0.001), as well as older age, white race, lower BMI and seronegativity, even when adjusting for symptom duration (all p< 0.03). Among the 203 on placebo with Day 0 RNA >= LLoQ, female sex had larger decreases in RNA at Day 3 vs male sex (difference in mean change: -0.8 log10 copies/mL (95% CI: -1.2, -0.4), p< 0.001) when adjusted for symptom duration and Day 0 RNA;this difference was also observed when evaluating the proportion with RNA < LLoQ at Day 3 (Risk Ratio (95% CI): 2.38 (1.11, 5.09)). Seropositivity at Day 0 was associated with higher probability of RNA < LLoQ at Days 3 and 7 (p< 0.001) in adjusted models. Seropositivity at Day 0 did not differ by sex. Conclusion. In this well characterized clinical trial cohort, shorter symptom duration, older age, white race, lower BMI and seronegativity were associated with higher RNA in early infection. Female sex and seropositivity were associated with earlier viral clearance. Further research is needed to determine if viral decay differences mediated by these host factors influence clinical outcomes.
ABSTRACT
Background. Symptoms during acute COVID-19 can limit daily activities and delay return to work and school. Little is known about the association between SARS-CoV-2 burden in either the upper airway or plasma and the duration of COVID-19 symptoms. Methods. ACTIV-2/A5401 is a platform trial for COVID-19 treatments in nonhospitalized symptomatic adults enrolled within 10 days of symptom onset. We included participants randomized to placebo from August 2020 to July 2021. Participants self-reported severity of 13 symptoms daily from day 0 (baseline) to 28 as Absent 0, Mild 1, Moderate 2, Severe 3;total symptom score was calculated as the sum of all scores. Anterior nasal (AN) and plasma SARS-CoV-2 RNA levels at day 0 were measured with a quantitative qPCR assay. The relationship between day 0 RNA and time to symptom improvement or resolution (first of 2 consecutive days of all symptoms improved or resolved from day 0, respectively) was evaluated using proportional hazards regression adjusted for time from symptom onset. Time to resolution of distinct symptoms was also assessed. Results. Among 570 participants randomized to placebo, median age was 48 years, 51% were female, and median time since symptom onset at baseline was 6 days;7% had prior COVID-19 vaccination. At day 0, AN RNA was detectable in 80% with a median of 4.1 log10 copies/ml (n=533, quartiles: 1.7, 6.0) and plasma RNA was detectable in 19% (91/476). Detectable plasma RNA at day 0, but not AN RNA, was associated with more severe symptoms at day 0 (2.4-point higher mean total symptom score, P=0.001). Both high AN (>=6 vs < 2 log10 copies/ml, adjusted hazard ratio [aHR] 0.63, P=0.001) and detectable plasma RNA (aHR 0.74, P=0.03) at day 0 predicted delayed symptom improvement. High AN RNA at day 0 also predicted a delay in symptom resolution (aHR 0.59, P=0.001). Both high AN RNA and detectable plasma RNA levels predicted delays in the resolution of cough and shortness of breath. Detectable plasma RNA also predicted delayed body pain resolution.
ABSTRACT
Background: The relationship between nasopharyngeal (NP) SARS-CoV-2 RNA, demographics and symptom characteristics in non-hospitalized persons with COVID-19 is not well described. Methods: ACTIV-2 is a phase 2/3 adaptive platform trial testing antivirals for SARS-CoV-2 in symptomatic non-hospitalized adults. We analyzed associations between NP quantitative SARS-CoV-2 RNA (Abbott m2000sp/rt) and COVID-19 symptomatology in 284 participants with both a NP swab and symptom diary prior to study intervention. The diary included 13 targeted symptoms and questions about overall severity of COVID-19 symptoms, each scored as none, mild, moderate, or severe (and very severe for overall severity) and general physical health (scored as poor, fair, good, very good, excellent). Wilcoxon tests were used to compare NP RNA levels between pre-defined groups. Spearman correlations, Jonchkeere-Terpstra trend tests, and linear regressions evaluated associations between symptom measures and NP RNA. Results: Participants were 49% female, 82% white, 9% black, and 27% Latinx. Median age was 46 years and 50% met the protocol definition of higher risk for COVID-19 progression (age ≥55 years and/or protocol-defined comorbidities);32% reported moderate and 5% severe symptoms. Median (Q1, Q3) time from onset of symptoms to NP swab/symptom assessment was 6 (4, 8) days. NP RNA was above the lower limit of quantification in 85%;median (Q1, Q3) was 5.4 (3.5, 6.8) log10 copies/mL. Higher RNA levels were associated with shorter symptom duration (median 6.5 vs 4.7 log10 copies/mL for ≤5 vs >5 days) but not total symptom score (Figure). Controlling for symptom duration, higher NP RNA levels were associated with better general physical health (p=0.02) and more severe body/muscle pain (p=0.04). No associations were observed with symptom severity (sum of scores or overall severity) or any other symptoms. There was no association between NP RNA and age or risk category for COVID-19 progression. Conclusion: In symptomatic outpatients, NP SARS-CoV-2 RNA levels were higher in persons with more recent symptom onset, but were not associated with symptom severity or risk for disease progression. The range of viral RNA shedding was remarkably similar across the range of symptom severity, suggesting symptom severity may not correlate with transmission risk or the potential to respond to antiviral therapy. Outpatient trials aimed at evaluating antiviral activity of new agents should focus enrollment on participants with recent onset of symptoms. (Figure Presented).
ABSTRACT
Background: Due to the substantial morbidity but low rates of hospitalization and death among outpatients with COVID-19, symptom outcome measures should be considered for primary efficacy assessment in phase 3 treatment trials. We analyzed potential measures utilizing the ACTIV-2 participant diary. Methods: Data from the first 95 participants in ACTIV-2 were included. All had symptomatic SARS-CoV-2 infection and received blinded bamlanivimab 7000 mg/placebo. The symptom diary was completed by participants prior to treatment (Day 0) and then daily for 28 days. It included 13 targeted symptoms scored as absent, mild, moderate, or severe, and a question about whether they had returned to pre-COVID-19 health. Without unblinding, 3 candidate symptom outcome measures were assessed: A) time to confirmed (2 consecutive days) absence of all targeted symptoms, B) time to all targeted symptoms confirmed to be mild or absent, and C) time to confirmed improvement in all targeted symptoms. Median time to outcome was estimated by Kaplan-Meier methods. Results: Of the 95 participants, 53% were female, 82% white, and 33% Latinx. Median age was 44 years;46% were age ≥55 years and/or had protocol-defined comorbidities. Median time from COVID-19 symptom onset to randomization was 6 days. Prevalence of each targeted symptom on Day 0 ranged from 6% vomiting to 87% fatigue. Candidate outcome B was met in median 2 days due to 29% of participants having only mild symptoms at Day 0. For candidate outcomes A and C, median time was 11 and 8 days, with 26% and 16%, respectively, not meeting the outcome by 28 days. These candidate outcomes (A and C) were associated with a participant's confirmed assessment of return to pre-COVID-19 health (Figure). For all measures, increasing the consecutive days required for confirmation from 2 to 3 or 4 had a modest impact on median time to the outcome being met, consistent with few participants experiencing relapsing symptoms. Conclusion: Outcomes based on symptom resolution (A) or improvement (C) are promising for evaluating COVID-19 treatment response, with good internal validity with self-assessment of return to pre-COVID-19 health. A valid symptom outcome measure may be preferred over hospitalization/death as a primary outcome for outpatient COVID-19 treatment trials as most participants achieve the outcome, increasing power to compare treatments, especially among participants who are at low risk for hospitalization/death.