Reduced Mortality in COVID‐19 Patients Treated With Inhaled Extracellular Vesicles Expressing CD24

ABSTRACT

Background: Severe COVID‐19, and other systemic infections, can trigger hyperinflammation leading to lung injury. EXO‐CD24, an inhaled extracellular vesicle therapy with anti‐inflammatory properties, has shown acute clinical benefit, but long‐term effects on survival remain unknown.

Methods: We compared the mortality in 35 patients with moderate‐to‐severe COVID‐19 treated with inhaled EXO‐CD24 on five consecutive days in 2020, versus 105 matched controls receiving standard care (1:3 ratio). This ‘real life’ study was a retrospective follow‐up extended to 4.6 years after infection, evaluating all‐cause mortality as well as health‐related quality of life in survivors (HRQoL).

Results: No deaths occurred in the EXO‐CD24 group (0/35) versus 14 deaths (13.3%) among the matched controls (p < 0.05). The survival difference was most marked in the first month but persisted over time. At the end of the follow‐up, theEXO‐CD24 patients reported significantly better HRQoL, including well‐being, physical function, and mobility.

Conclusions: Repeated inhalations with EXO‐CD24 is associated with improved survival and HRQoL up to 4.6 years post‐treatment. These findings suggest durable benefits and support further evaluation of EXO‐CD24 as an immunomodulatory therapy in severe inflammatory conditions.

Trial Registration: ClinicalTrials.gov identifier: NCT04657497

1. Introduction

SARS‐CoV‐2, the virus that spread globally in 2020, causes infections ranging from asymptomatic illness to fatal disease. Mortality was most often driven by severe lung injury leading to respiratory failure, although multiple organs could be affected. Many survivors continue to experience persistent symptoms for years after infection, with detrimental effects on long‐term health‐related quality of life (HRQoL) (Nalbandian et al. 2021; Yong 2021; Centers for Disease Control and Prevention 2022). Mechanisms associated with severe COVID‐19 include a dysregulated immune response and hyperinflammation, or ‘cytokine storm,’ which drives acute lung injury and tissue damage. These insights underscore the need for therapies that curb excess inflammation while preserving antiviral immune defense mechanisms.

CD24 is a membrane‐bound immunoregulatory protein that binds Siglec‐10 to dampen damage‐associated molecular pattern (DAMP) responses while preserving pathogen‐specific associated immunity (Chen et al. 2009; Barkal et al. 2019). Based on this pathway, EXO‐CD24, an extracellular vesicle (EV) therapy overexpressing CD24, was developed to treat severe inflammation. In 2020, a first‐in‐human Phase Ib/IIa trial tested inhaled EXO‐CD24 in 35 hospitalized patients with moderate‐to‐severe COVID‐19 (Shapira et al. 2022). The 5‐day escalating‐dose treatment was found to be safe and well tolerated, and was linked to rapid improvements, including reduced oxygen needs, fast radiographic recovery, short hospitalization, and normalized cytokine profiles. A subsequent multicenter Phase IIb trial in Greece (n = 91) confirmed favorable safety and efficacy signals (Grigoropoulos et al. 2024), findings further supported in a compassionate‐use cohort (Green et al. 2023).

Although the earlier studies established the safety and acute clinical benefit of EXO‐CD24, the long‐term impact of early immune modulation in COVID‐19 remained unknown. Therefore, we hypothesized that EXO‐CD24 treatment during acute infection would convey clinically important benefits, including reduced mortality and improved long‐term HRQoL. To test this, we conducted a retrospective follow‐up of 35 patients from the first clinical trial and compared them with 105 matched controls of similar age, comorbidities, and disease severity who received standard care (Al‐Aly et al. 2021).

2. Materials, Methods, and Participants

2.1. Study Objective

The primary objective was to determine whether EXO‐CD24 treatment is associated with improved outcomes, including reduced mortality and better long‐term HRQoL. Outcomes were compared between the 35 patients treated in 2020 with 105 historical controls (1:3 ratio) matched for age, sex, comorbidities, disease severity, and viral strain, who received standard of care (SOC) during the same period.

2.2. Study Design

For comparison with the Phase Ib/IIa trial results, a historical cohort of 105 control patients was assembled at Tel Aviv Medical Center (TASMC), Israel, with follow‐up from 6 June 2020, to 30 June 2025. Controls were matched to the intervention group by age, sex, baseline disease severity, comorbidities, and SARS‐CoV‐2 variant. The primary endpoint was all‐cause mortality, and secondary endpoints focused on long‐term morbidity such as HRQoL.

2.3. Inclusion Criteria

All 35 participants from the original Phase Ib/IIa trial and their 105 matched controls were included in this real‐world analysis, with inclusion criteria consistent with those previously reported (Barkal et al. 2019). Mortality status was continuously updated through the Israeli Ministry of Health (IMOH) registry. Longitudinal follow‐up of the EXO‐CD24 group began at trial enrollment, while surviving controls were identified after recovery and contacted every 6 months. Verbal informed consent was obtained from all living participants during phone interviews, as approved by the TASMC Institutional Review Board (IRB).The primary endpoint was all‐cause mortality over a the follow‐up period. Secondary endpoints included long‐term morbidity, incidence of late adverse events potentially related to immune dysregulation, patient‐reported overall well‐being compared with pre–COVID‐19 baseline, and health‐related quality of life (HRQoL). HRQoL was assessed using the validated Hebrew version of the EQ‐5D‐5L (EuroQol 5 Dimensions, 5 Levels) questionnaire. Patient‐reported outcome measures (PROMs) were collected on a standardized 5‐point scale and analyzed alongside EQ‐5D‐5L data to quantify HRQoL. The EQ‐5D‐5L consists of a descriptive system covering five dimensions of health (mobility, self‐care, usual activities, pain/discomfort, and anxiety/depression) and a visual analogue scale rating overall health status.

2.4. Data Collection

Information on symptoms, hospitalizations, new diagnoses, and quality of life was obtained through structured, physician‐administered telephone interviews and corroborated with medical records when available. Mortality data were verified in the national IMOH registry.

2.5. Ethics and Compliance

The study protocol was approved by the Tel Aviv Sourasky Medical Center (TASMC) Institutional Review Board (IRB). All procedures were conducted in accordance with institutional and national research committee standards and with the principles of the Declaration of Helsinki. Data collection and analysis complied with applicable local data protection regulations.

2.6. Statistical Analyses

Baseline characteristics were summarized with descriptive statistics. Mortality data were obtained from the Israeli Ministry of Health (IMOH) registry. Survival was analyzed using Kaplan–Meier curves with log‐rank testing. Fisher's exact test was used for mortality comparisons due to the small sample size and zero events in the treatment group, while chi‐square tests were applied to other categorical variables. Logistic regression and Cox proportional hazards models adjusted for potential confounders. Analyses were exploratory; two‐sided p < 0.05 was considered statistically significant.

3. Results

In terms of baseline characteristics, the groups were very similar (Table 1). There are no statistically significant differences in mean age, gender and comorbidities between the groups, besides a higher rate of obstructive lung disease in the intervention group. All patients became ill and were hospitalized during the same time period, thus neutralizing the continuous change in COVID‐19 disease severity, as well as the evolution in standard care and its increased effectiveness. The patients in both groups met the criteria for the original study in terms of disease severity (inflammation indices, hospitalization, and oxygenation). Laboratory indicators did differ significantly, the ratio of neutrophils to leukocytes (NLR), which has been demonstrated as an effective predictor of severe disease development (22) was also similar between the two groups. There was a tendency of the active treatment group having slightly higher prevalence of obstructive lung disease, compared to the matched control group.

TABLE 1.

Baseline characteristics of the treatment versus matched control groups.

Characteristic		Intervention N (%) or Mean (SD)	Control N (%) or Mean (SD)	Test statistic	p value
Age at visit (Years)		57.5 (11.46)	55.6 (12.3)	T = 0.492	0.624
Gender (%)	Female	12 (34.3)	38 (36.2)	X 2 = 0.041	0.839
	Male	23 (65.7)	67 (63.8)
Diabetes (%)	Yes	6 (17.1)	25 (23.8)	X 2 = 0.677	0.411
	No	29 (82.9)	80 (76.2)
Obesity (%)	Yes	8 (22.9)	34 (32.4)	X 2 = 1.134	0.287
	No	27 (77.1)	71 (67.6)
CRF (%)	Yes	1 (2.9)	10 (9.5)	X 2 = 1.611	0.204
	No	34 (97.1)	95 (90.5)
Obstructive lung	Yes	7 (20.0)	9 (8.6)	X 2 = 3.387	0.066
Disease (%)	No	28 (80.0)	97 (91.4)
IHD (%)	Yes	4 (11.4)	14 (13.3)	X 2 = 0.085	0.771
	No	31 (88.6)	90 (86.7)
CHF (%)	Yes	0 (0.0)	6 (5.7)	X 2 = 2.090	0.148
	No	35 (100.0)	99 (94.3)
Remdesivir (%)	Yes	21 (60)	77 (73.3)	X 2 = 3.233	0.072
	No	14 (40)	24 (26.7)
Dexamethasone (%)	Yes	29 (82.86)	102 (97.1)	X 2 = 4.060	0.044
	No	6 (17.14)	3 (2.9)
Actemra (%)	Yes	4 (11.4)	16 (15.2)	X 2 = 0.311	0.577
	No	31 (88.6)	88 (84.8)
NLR > 6.82 (%)	Yes	12 (34.3)	48 (45.7)	X 2 = 1.400	0.237
	No	23 (65.7)	57 (54.3)
Baseline CRP (mg/L)		130.96 (87.21)	133.5 (79.3)	T = 1.21	0.227
Baseline WBC (mg/L)		8.06±3.95	7.35±3.24	T = 1.3	0.25

Abbreviations: CHF, chronic heart failure; CKD, chronic kidney disease; CRP, C reactive protein; NLR, neutrophil to leukocyte ration; Obesity, BMI above 30; SD, standard deviation.

*p value of <0.05 indicating statistical significance.

3.1. Efficacy Endpoints

The primary outcome was overall mortality. No deaths occurred in the EXO‐CD24 group (0/35, 0%) compared with 14 deaths in the control group (14/105, 13.3%). Fisher's exact test, applied due to zero events in the treatment group, showed a statistically significant difference between groups (p ≈ 0.03). Kaplan–Meier cumulative survival curves for the intervention and matched control groups are shown in Figure 1. Survival curves diverged immediately after COVID‐19 diagnosis, with half of control deaths (7/14) occurring in the first month, and the additional seven at later time points. No deaths occurred in the EXO‐CD24 group, whose survival curve remained flat throughout follow‐up. The survival distributions between the intervention and control groups differed significantly by log‐rank test (Mantel–Cox), χ ²(1) = 4.99, p = 0.026.

FIGURE 1.

Kaplan–Meier cumulative survival curves for the EXO‐CD24 group (blue) and historical matched controls (red). Censored observations are marked with “+”. The EXO‐CD24 group maintained 100% survival throughout the follow‐up period, whereas survival in the control group declined progressively over time, with a total of 14 deaths (13.3%).

Subgroup analysis within the matched control group suggested a higher mortality in males than females (12/72, 16.7% vs. 2/33, 6.1%). However, although a numerical difference was observed, it was not statistically verified. No deaths occurred in the EXO‐CD24 group, precluding any subgroup analysis for those patients.

The secondary endpoint was HRQoL assessed by EQ‐5D‐5L (Table 2). Significant group differences (p < 0.05) were observed in mobility (1.1 ± 0.1 vs. 1.5 ± 0.1) and usual activities (1.1 ± 0.1 vs. 1.5 ± 0.1), with EXO‐CD24 patients reporting fewer limitations. EQ‐VAS scores were higher in the intervention group (91.5 ± 2.7 vs. 83.7 ± 2.3), though they recalled more severe illness during acute infection (3.5 ± 0.8 vs. 11.8 ± 1.7). Overall, however, at the 5‐year follow‐up, most scores across domains indicated no or slight problems, reflecting little morbidity overall in the survivors.

TABLE 2.

EQ‐5D‐5L questionnaire results in the EXO‐CD24 and historical cohort groups.

	EQ‐5D descriptive system					EQ visual analogue scale (VAS) – Overall assessment of health on a VAS from 1 to 100
	Mobility	Self‐care	Usual activities (work, studies, housework, family or leisure activities)	Pain/discomfort	Anxiety/Depression	Current 1–100 VAS score	Recall of 1–100 VAS score during COVID‐19
EXO‐CD24 intervention (n = 35)	1.1±0.1*	1.1±0.1	1.2±0.1*	1.5±0.2	1.2±0.1	91.5±2.7*	3.5±0.8*
Historical control (n = 105)	1.5±0.1	1.2±0.1	1.5±0.1	1.7±0.1	1.1±0.1	83.7±2.3	11.8±1.7

Note: Domain scores are shown as mean ± SD.

*p value < 0.05.

3.2. Safety Endpoints

No serious adverse events (SAEs) or other adverse events (AEs) related to the drug were reported through the follow‐up period in the treated group (Barkal et al. 2019).

4. Discussion

This long‐term follow‐up study demonstrates that inhaled EXO‐CD24 given to patients with lung injury associated with acute moderate‐to‐severe COVID‐19 results in sustained clinical benefits. Over nearly 5 years of follow‐up, no deaths occurred among the 35 EXO‐CD24–treated patients, compared with 14 deaths (13.3%) in 105 matched historical controls. As expected, most deaths in the control group occurred within the first month of COVID‐19 infection, yet additional mortality accumulated over the observation period. In contrast, EXO‐CD24‐treated patients remained protected, with no late deaths recorded. Treatment was also associated with significantly better long‐term health‐related quality of life, particularly in mobility, usual activities, and overall well‐being.

The observed protective effect on COVID‐19‐related mortality is notable for several reasons. Few therapies have demonstrated durable survival benefits. IL‐6 inhibitors (REMAP‐CAP Investigators et al. 2021), Janus kinase (JAK) inhibitors (Limen et al. 2022), and possibly corticosteroids (Chaudhuri et al. 2021) have been suggested to reduce short‐term mortality in severe COVID‐19, but these studies are largely confined to the acute phase of the disease, with no long‐term observation periods reported. Antiviral agents such as nirmatrelvir/ritonavir (Paxlovid) have also reduced mortality in both the acute and post‐acute phases, including a 32% decrease in long‐COVID–related mortality (Chuang et al. 2025) and a 63% reduction in 1‐year mortality among patients with chronic kidney disease (Strohbehn et al. 2024), yet sustained multi‐year benefit remains undocumented. Like EXO‐CD24, mesenchymal stem cell‐derived EVs have demonstrated survival benefit in acute COVID‐19, reinforcing the potential of EV‐based therapies in hyperinflammatory conditions, although long‐term effects are still to be established (Lightner et al. 2023; Zamanian et al. 2024). Thus, while several therapies can reduce acute mortality in COVID‐19, additional evidence for sustained long‐term survival benefits are needed, underscoring the significance of the durable protection observed with inhaled EXO‐CD24.

The mechanism by which EXO‐CD24 protects against mortality is not fully elucidated. Still, it is possible that the inhaled drug rapidly restores immune homeostasis via the CD24–Siglec checkpoint, thereby interrupting a destructive inflammatory cycle before permanent tissue injury takes hold. It is thus likely that the EXO‐CD24 can have the ability to prevent tissue injury, such as ARDS, following COVID‐19 infection. This tissue repair is associated with an excessive innate immune response, which suggests that EXO‐CD24 may be effective in other types of excessive inflammatory responses.

Beyond survival, the observation of improved HRQoL several years after treatment suggests that EXO‐CD24 may influence long‐term recovery trajectories in COVID‐19 survivors. Persistent symptoms, collectively termed long‐COVID, remain a major public health burden and are thought to arise, at least in part, from sustained immune dysregulation and tissue damage (Sudre et al. 2021). By dampening pathological inflammation while preserving antiviral defense, EXO‐CD24 could plausibly mitigate the biological processes that lead to chronic post‐viral sequelae. These findings therefore raise the possibility that targeted immunomodulation during acute infection not only reduces short‐term mortality but also alters the risk of long‐term disability, a therapeutic goal largely unmet by current COVID‐19 treatments. EXO‐CD24 is most logically positioned as an early intervention to prevent ARDS. Yet its therapeutic window appears broad, with no reduction in efficacy was observed even when treatment is initiated in slightly later stages of disease.

Baseline characteristics were largely comparable between groups, with no significant differences in age, sex, or comorbidities, except for a numerically higher prevalence of obstructive lung disease in the intervention group. All patients were hospitalized during the same time period, minimizing bias from evolving disease severity or improvements in standard care. Both groups met the original study's inclusion criteria regarding disease severity (inflammatory indices, hospitalization, and oxygenation). Laboratory indicators were also similar, including the neutrophil‐to‐lymphocyte ratio (NLR), a recognized predictor of severe disease progression. Importantly, despite the higher prevalence of obstructive lung disease in the intervention group, no deaths occurred, which is notable since pre‐existing obstructive lung disease is a well‐documented risk factor for COVID‐19 mortality (Peckham et al. 2020; Rabbani et al. 2021; Scully et al. 2020).

Several limitations should be acknowledged. This was a retrospective, non‐randomized study with inherent risk of selection bias, despite 1:3 matching for key prognostic variables. The control group was derived from historical data, and unmeasured confounders cannot be excluded. Secondary endpoints such as HRQoL were based on patient‐reported outcomes and may be subject to recall or reporting bias, but showed differences despite the matched control group having lost lives during the observation period. Nonetheless, use of Fisher's exact test supports the robustness of the survival analysis, even with the small sample size and absence of events in the treatment group.

By attenuating DAMP‐driven inflammation without compromising pathogen‐directed immunity, EXO‐CD24 represents a promising therapeutic strategy for conditions where hyperinflammation is the primary driver of pathology. Its potential application extends to bacterial sepsis, severe trauma, and perioperative inflammatory syndromes. Further, the inhaled formulation, which was used in this study, is particularly suited for sterile inflammatory lung diseases such as certain phenotypes of ARDS, COPD, and non‐allergic asthma. However, these applications remain speculative, as evidence beyond COVID‐19 is currently lacking. Thus, robust, indication‐specific randomized controlled trials are needed to definitively establish its broader efficacy beyond COVID‐19, notwithstanding the very favorable preliminary safety data found until now.

This study provides one of the longest follow‐ups of an investigational immunomodulatory therapy given during acute COVID‐19, demonstrating durable improvements in survival and HRQoL. Lung injury caused by systemic infections such as COVID‐19 and sepsis remains a major global health challenge, and therapies like EXO‐CD24, with a beneficial safety profiles and targeted anti‐inflammatory activity, may offer a new approach in managing hyperinflammatory diseases. These findings support continued development of EXO‐CD24 as a safe and scalable immunomodulatory therapy with potential applications in both acute and chronic inflammatory conditions.

Author Contributions

Shiran Shapira: conceptualization, methodology, investigation, formal analysis, data curation, writing – original draft, writing – review and editing, visualization, funding acquisition, supervision. Guy Vishnevsky: data curation, writing – original draft. Hadas Yaakobi: investigation, data curation. Gil Shenberg: investigation, data curation. Mays Motlaq: project administration. Eliezer Liberman: investigation, data curation. Aminadav Kalisky: investigation, data curation. Gadi Benari: writing – review and editing. Sotirios Tsiodras: writing – review and editing. Nimrod Adi: writing – review and editing. Jan Lötvall: writing – original draft, writing – review and editing. Nadir Arber: conceptualization, methodology, investigation, formal analysis, data curation, writing – review and editing, writing – original draft, supervision, funding acquisition.

Ethics Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (IRB) of Tel Aviv Sourasky Medical Center. All participants provided written informed consent prior to inclusion in the study.

Conflicts of Interest

Drs. Shapira and Arber are employees of Nano24 and hold stock options in the company. Kazanov is an employee of Nano24. JL consults for ExoCoBio INC (South Korea), and is the founder of Exocure Sweden AB and Nexos Biosciences AB. The other co‐authors declare no competing interests.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.