A Phase II Randomized Multi-Center Study to Evaluate the Safety and Efficacy of Infusion of Memory T Cells As Adoptive Therapy in Coronavirus Pneumonia and/or Lymphopenia (RELEASE NCT04578210)

BACKGROUND There are currently no specific anti-viral treatments for COVID-19 patients. Nonspecific treatments such as dexamethasone had become the standard of care (SoC) for hospitalized patients. SARS-CoV-2 induced lymphopenia is a biomarker of disease severity usually presented in hospitalized patients. We have previously demonstrated in phase 1 clinical trial that infusion of CD45RA^- memory T cells from a convalescent donor containing SARS-COV-2 specific T cells is safe and feasible. Now we have performed a phase 2 randomized multicenter clinical trial where we evaluated the safety and efficacy of the infusion of memory T cells vs the SoC treatment. We hypothesize that the pool of memory T cells will increase in these patients and the restoration of total lymphocytes will help to clear the infection.

METHODS Patients were enrolled from January 2021 to January 2022 in Spain, at University Hospital La Paz and Hospital de Emergencias Enfermera Isabel Zendal in Madrid and Hospital Clínic in Valencia. Eighty-six patients were randomized into SoC arm and SoC plus the infusion of 1x10⁶/kg memory T cells containing SARS-CoV-2 specific T cells from convalescent donors. Analysis was completed with 81 patients (Figure 1). Eligible donors were chosen based on the expression of SARS-CoV-2 specific response within the CD45RA^- memory T cells and frequency of HLA epitopes in the Spanish population. Cells from 4 unvaccinated donors were biobanked. Primary outcome was defined as proportion of participants in each group with normalization of fever and oxygen saturation and lymphopenia recovery through day 14. Secondary outcomes were 1) time to normal level of lymphocytes, 2) patients showing clinical improvement at day 7 according to the investigator, 3) patients receiving a second cycle, 4) days to first negative SARS-CoV-2 PCR after infusing memory T cells, 5) incidence of treatment-related adverse events, 6) days of hospitalization, 7) days to discharge or to a NEWS of ≤ 2 and maintained for 24 hours, 8) Time to improvement by one category on a 7-point ordinal scale, 9) subject clinical status (on a 7-point ordinal scale) at day 14, 10) patients requiring intensive care unit (ICU), 11) all-cause mortality at day 28. Exploratory outcomes were defined as donor chimerism and immune lymphocyte reconstitution.

RESULTS Of the analyzed participants median age was 58.5 (50-65.25) years in the SoC group and 56 (50-65) in the infusion arm (p=0.643). 72.83% of patients were men, and 27.16% women (p=0.221). No statistically differences were found at baseline between the two groups including body mass index, vaccination status at the time of enrollment, comorbidities and concomitant treatments. Median time from symptom onset to randomization was 9 (7-10.25) days, and from randomization to discharge was 9 (6-16) days. Two patients died in SoC group and none in the experimental arm (p= 0.19) on day 28. Total mortality accounted for 4 deaths in the SoC arm and 1 in the experimental arm (p=0.17).

Primary outcome met the pre-specified requirements for efficacy. 74.1% of patients in the infused arm and 40% in SoC arm showed recovery according to the protocol (p=0.01). Also, 87.1% of patients in the infusion arm presented lymphocyte recovery on day 14 (p=0.02).

The percentage of patients showing lymphocyte recovery on days 3 and 14 was higher in the experimental arm (p= 0.04). No statistically significant differences were found in the other secondary outcomes or pro inflammatory parameters at enrollment or during time.

The proportion of patients showing total lymphocyte recovery was higher in the experimental arm on days 3 and 14 (p=0.004, p=0.048). Also on day 7, the same group showed an increase in NK cells (p=0.04). In T cell subsets we observed a mean of the percentage of CD8RA⁺ cytotoxic T cells on day 7 of 50 (42-59) in the infused arm and 37 (20-46) in the SoC (p<0.0001). Donor microchimerism was observed for 3-4 weeks after infusion. There were no treatment-related adverse events.

CONCLUSIONS To the best of our knowledge this is the first clinical study evaluating the safety and efficacy of the infusion of CD45RA^- memory T cells containing a pool of SARS-COV-2 specific T cells from convalescent donors in hospitalized COVID-19 patients. Our results met the established primary outcome for patient recovery. Further outcome improvements can be achieved by optimizing the therapy and selecting the best donors based on current vaccination and infections.

Soria:Celgene: Consultancy; Gilead: Consultancy; Takeda: Consultancy.